# Amass — full site content > Concatenated markdown of every static page on amass.tech, regenerated at each build. The curated index (with per-page .md links) is at https://amass.tech/llms.txt. # Amass — Scientific intelligence you can verify Answers are only as useful as the evidence behind them # Scientific intelligence you can verify. Amass searches papers, clinical trials, drugs, genes, patents, regulatory records and the live web. Every answer stays connected to its sources, so you can inspect the evidence behind it. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) Free plan, $0 forever. No card required. [01App Ask a research question and watch it work: every claim cited, and what it read but didn’t cite shown too.](#cr-panel-app) [02MCP Amass inside the tools you already use: Claude, ChatGPT, Claude Code and Codex, over MCP.](#cr-panel-mcp) [03API Build on the engine: typed, cited responses you can program against.](#cr-panel-api) [See the App →](https://amass.tech/app) One question, three ways in: the app with its trace and citations, an assistant calling the Amass tool over MCP, and the API returning the record. Each answer cites the ID of the record behind it, such as a PMID, NCT ID or patent number, in the Amass app, in Claude and ChatGPT over MCP, or from the API. ## A fluent answer is not a checkable one. Amass returns each answer with its sources. Each claim carries an identifier that opens the record it came from: a DOI, an NCT ID, a ChEMBL ID, a patent number or a regulatory application number. peer-reviewed papers 43M+ BioMedCore scholarly papers 233M+ ScholarCore\* patent filings 170M+ PatentCore one schema FDA + EMA RegulatoryCore drugs & molecules 22K+ DrugCore clinical trials 1.2M+ TrialCore human genes 43K+ GeneCore real-time web sources Live WebCore\* \* In the app; the API and MCP serve the other six. 1. **01**ScatterYour sources sit on separate platforms 2. **02**DriftThe same question gets different answers over time 3. **03**ConvergeAmass traces each claim to the record it came from 4. **04**CiteEvery answer shows its sources, so the team shares one picture Research holds the answer ## Evidence is scattered across teams, tools and documents. ## Reasoning drifts across data, teams and time. ## Amass resolves it to the record. Every claim cited to source. One question about a medicine can touch a journal article, a trial registry, a patent and a regulatory label, then the slides and emails a team writes about them. None of them links to the others. Each answer starts from whichever documents someone found. A slide repeats one reading of a result, an email another, and the readings drift as they are copied forward. None says where its figure came from. Behind every claim sits a record, held in the \*Core for that kind of source. A slide or an email is not a record, so Amass reads it but never cites it. JournalBioMedCore [10.1056/NEJMoa2413258](https://doi.org/10.1056/NEJMoa2413258) RegistryTrialCore [NCT04822181](https://clinicaltrials.gov/study/NCT04822181) PatentPatentCore [US-2025375504-A1](https://patents.google.com/patent/US20250375504A1/en) LabelRegulatoryCore [NDA 215256](https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=215256) SlideQ3 review, slide 12 not a source EmailRe: MASH readout, v7 not a source Each reference opens its record. What was read but not cited stays on a list of its own, so the team works from one picture it can check. The platform ## One evidence base for researchers, agents and apps. One record, served three ways: as the app writes it, as the API returns it and as an MCP tool call fetches it. The app searches all eight \*Cores; the API and the MCP serve six. 1. Grounded research across every surfaceUse Amass in the app, through agents, or directly via API and MCP. 2. One engine over eight \*Cores\*Cores for papers, trials, drugs, genes, patents, and FDA + EMA regulatory records. 3. Every claim traceable to sourceAnswers, workflows and outputs stay linked to the underlying evidence. Researchers Apps & Agents Developers 8 \*Cores App MCP API Amass \*Core architecture BioMedCore 43M+ ScholarCore 233M+ PatentCore 170M+ RegulatoryCore 7K+ DrugCore 22K+ TrialCore 1.2M+ GeneCore 43K+ WebCore Live Provenance & citations External data providers PubMedLiteraturePMCLiteratureCrossrefLiteratureUSPTOPatentsEPOPatentsWIPOPatentsSureChEMBLPatentsFDARegulatoryEMARegulatoryClinicalTrials.govTrialsWHO ICTRPTrialsHGNCGenes & moleculesEnsemblGenes & moleculesUniProtGenes & moleculesChEMBLGenes & moleculesWebWeb Connect your own data SharePointYour dataOneDriveYour dataGoogle DriveYour dataDatabricksYour dataS3 / BlobYour dataNetwork drivesYour data [See the three surfaces →](https://amass.tech/product) MCP ## Inside the assistant you already use. Amass runs inside Claude, ChatGPT, Claude Code and Codex over the Model Context Protocol (MCP). The assistant calls an Amass tool, and the answer comes back with the identifiers. [Set it up →](https://amass.tech/mcp) ClaudeChatGPTClaude CodeCodexAmass connected Which Lp(a)-lowering programs are in phase 3? amass`search_amass_trialcore_records`Completed `query: "lipoprotein(a) cardiovascular outcomes"``phase: "PHASE3"` [NCT04023552](https://clinicaltrials.gov/study/NCT04023552)Lp(a)HORIZON · COMPLETED · Novartis Pharmaceuticals [NCT05581303](https://clinicaltrials.gov/study/NCT05581303)OCEAN(a)-Outcomes · ACTIVE\_NOT\_RECRUITING · Amgen Two phase 3 programs: Lp(a)HORIZON (NCT04023552) has completed, and OCEAN(a)-Outcomes (NCT05581303) is active and no longer recruiting. Reply… An assistant calling Amass’s TrialCore search tool, and the two trial records it hands back. API ## The same record, as a response. Build on the engine with one REST API across six cross-linked \*Cores. Each response is typed and carries the identifiers of its sources. GET`https://api.amass.tech/api/v1/cores/trialcore/records?query=NCT04023552&limit=1`Authorization: Bearer •••• 200 OKapplication/json {"data": \[{"nctId": "[NCT04023552](https://clinicaltrials.gov/study/NCT04023552)","acronym": "Lp(a)HORIZON","phase": "PHASE3","overallStatus": "COMPLETED","sponsorName": "Novartis Pharmaceuticals","enrollment": 8323,"completionDate": "2026-07-16","referencesBiomedCore": \[ "AMBC\_…", … \],}\]} One trial record from the documented TrialCore records path, as the API returns it. [API docs →](https://platform.amass.tech/documentation/getting-started/overview) ## Used by people at > Amass gives our investment team one place to validate science, patents, and trials, with citations we can stand behind in committee. Luis Toledo Senior Scientist, Novo Holdings SeedLab ## Your research, cited to source. Start free in the app or with an API key. Enterprise deployments add an SLA, single-tenant hosting and onboarding support. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # About About Amass # Built for scientific work that runs on evidence. The evidence behind a scientific decision sits in too many systems. Amass connects papers, trials, patents and regulatory records in one scientific intelligence layer, so your team can search, verify and reuse it without losing the source trail. Mission ## Scientific evidence you can trace and reuse. Life science decisions are only as good as the evidence behind them. That evidence is scattered across papers, patents, trials, regulatory sources and internal documents. Amass brings it into one layer. Every answer links back to its source, so researchers, investors, developers and strategy teams work from the same evidence. Your people make the call. Amass supplies the evidence underneath. 43M+papers in BioMedCore 8\*Cores in the app 2022Founded CopenhagenDenmark How we got here ## From biomedical search to scientific intelligence infrastructure Amass started from one problem: life science teams make important decisions from evidence scattered across many sources. We began with biomedical retrieval: searching the literature with precision and a source trail. Amass has since grown into a scientific intelligence layer that connects evidence sources, structures them into workflows and links every answer to its source. Teams use it for research, due diligence, competitive intelligence, portfolio monitoring, target evaluation and scientific strategy. The \*Cores ### Eight data layers underneath - BioMedCoreBiomedical literature and citations. - ScholarCoreScholarly literature across every discipline. - PatentCorePatents, claims, families and ownership. - RegulatoryCoreFDA, EMA and regulatory documents. - DrugCoreDrugs, molecules and modalities. - TrialCoreClinical trials, sponsors and endpoints. - GeneCoreHuman genes, targets, and druggability. - WebCoreReal-time company, market and conference news. App, API and MCP The app has all eight \*Cores. The API and the MCP serve six of them to your own code and AI tools. How we work ## How Amass handles the evidence ### Sources before answers Amass retrieves the source material first, then structures it, synthesizes it and reasons over it. ### Built for life science data Publications, patents, clinical trials, regulatory sources, web data and internal documents, worked on together. ### Traceable by default Every answer, workflow and output links back to its source, so your team can review, verify and reuse the evidence. Founders ## Scientific depth and commercial execution ### Alexander Junge Co-Founder & CTO Bioinformatics PhD and former Novo Nordisk data scientist. Built the Amass data engine from the ground up, combining biomedical retrieval, scientific data infrastructure and applied AI for life science workflows. PhD BioinformaticsNovo Nordisk18,000+ citationsAI/MLScientific retrieval ### Henrik Jensen Co-Founder & CEO Commercial and organizational lead with experience scaling digital products, fundraising, partnerships and go-to-market. Previously built and exited product-led companies and advised life science and technology organizations. CEO / GTMIPO experienceExit experienceVC / fundraisingLife science strategy Team ## People who have worked inside the problem ### Jörn Petersen AI & Data Engineer Founding engineer behind Novo Nordisk's BraiNN biomedical knowledge graph. Builds the data pipelines, retrieval systems and scientific infrastructure behind Amass. Novo Nordisk BraiNNBiomedical knowledge graphsIBM Watson Health ### René Bischoff Platform and Infra Engineer Full-stack and infrastructure engineer. First engineer at Pleo and built their infrastructure from the ground up. Builds the platform layer that makes Amass reliable in production. Platform EngineerInfrastructureAPI and MCP ### Lluís Colomer Coll AI Engineer Leads development of Amass agents and applied AI workflows. Focused on model orchestration, research workflows and production-grade scientific agents. MSc Human-Centered AIAmass Agents lead ### Emil Cronval GTM Engineer Builds the bridge between product, operations and go-to-market. Owns automation, customer workflows and internal systems that help Amass move faster. AI GTMOperationsAutomationSales workflows ### Heidi Noddeland Business Development Life science commercial strategy and partnerships. PhD in pharmaceutical sciences with experience in pharma, biotech and innovation consulting. PhD Pharmaceutical SciencesLEO PharmaBioInnovation / consultingPartnerships ### Niels Andersen Product & UX Strategy Shapes product direction and user-facing workflows. 20 years of product strategy and design across enterprise and startup environments. Product strategyUXEnterprise softwareLife science workflows ### Kris Modig Scientific Solutions, Enterprise Growth Lead Helps life science organizations put the Amass scientific data layer to work in their research and decision-making. Five years bridging science and data in antibody and biologics discovery, most recently at Benchling and before that at PipeBio. BenchlingPipeBio We're hiring ### ML / Research Engineer Work on retrieval, grounding and scientific reasoning across eight \*Cores, from 43M+ papers to 170M+ patents. You know ML well and ship to production. Copenhagen, on-siteFull-time [See the role→](https://amass.tech/careers) We're hiring ### Platform Engineer Scale the infrastructure that serves the \*Cores and the API to research teams. Backend, data and distributed systems. Copenhagen, on-siteFull-time [See the role→](https://amass.tech/careers) Open application ### Don't see your role? If you work in life science, infrastructure or AI, write to us. All disciplines [Get in touch→](mailto:hello@amass.tech) Board ## The board behind the strategy The board oversees governance and strategy. ### Dr. Christian Thirion Chairman Founder and MD of SIRION Biotech GmbH, a Munich-based biotech firm recognized as a leader in viral vector development and cell modeling for preclinical research (sold to Revvity). Also serves as board member at Lysomab and Hervolution. ### Tommy Ahlers Board member Partner at Giant Ventures. Former Danish Minister for Science and Higher Education and MP. Founded/built ZYB and Podio (exit to Vodafone and Citrix) and chair of Peakon and Lix. ### Bodil Sidén Board member General Partner at KOST Capital, an early-stage VC investing in food and bioeconomy-driven innovation. Former Partner at blq invest, backing early-stage ventures like Ingrid, Bower and Hydda. Backed by ## Our investors Amass is backed by bio, tech and pharma investors. ## Build scientific intelligence infrastructure with us. If you want to work on the evidence behind scientific decisions, see the open roles or write to us. [See open roles →](https://amass.tech/careers)[Get in touch →](mailto:hello@amass.tech) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Life Science API — Six *Cores behind one REST API API # Six \*Cores behind one REST API. One endpoint pattern searches biomedical literature, clinical trials, drugs, genes, patents, and FDA and EMA regulatory records. Each record comes back as typed JSON with its source identifiers (PMID, NCT ID, ChEMBL ID) and links to related records in the other \*Cores. Papers and trials update daily; drugs, genes and FDA and EMA records weekly. [Get an API key](https://platform.amass.tech/api-keys)[Read the docs](https://platform.amass.tech/documentation/getting-started/overview) [NewPatentCore is live: 170M+ patents, cross-linked.](https://amass.tech/core#patentcore) bash
1
$ curl \
2
 "https://api.amass.tech/api/v1/
3
 cores/drugcore/records?
4
 query=semaglutide&limit=1
5
 &include=referencesRegulatoryCore" \
6
 -H "Authorization: Bearer amass_YOUR_KEY"
7
8
→ 200 OK 1 record
responsedata\[0\] { "amassId": "AMDC\_G6w6ENSwfDNoeMc7x0OAzjGBlZj" "chemblId": "CHEMBL2108724" "name": "SEMAGLUTIDE" "tradeNames": \["Ozempic", "Rybelsus", …, "Wegovy"\] "drugType": "PROTEIN" "maxClinicalStage": "APPROVAL" "referencesRegulatoryCore": \["AMRC\_Ky54yeUZWbtHapyTC5q8z2GRMWO", …\] … } Join life-science AI builders on Discord [Join now](https://discord.gg/sEGaBHMhWa) Demo ## The \*Core launch film The six \*Cores the API serves, and the MCP, app and agents built on them. Builders ## What people build with it Founders, fellows, and hackathon teams who pointed their agents at the API and shipped. > We'd tried scraping patent data ourselves and it was a nightmare. With PatentCore we built a full freedom-to-operate landscape of our space in a couple of days. > The API is really great — I've used it personally for my structural-biology workflows. BioMedCore surfaced epitope annotations we couldn't find in structured databases like UniProt. > Of everything we built on during the hackathon, Amass was the most valuable API. I built a landscape around patents and trials — very, very valuable for understanding where the field is going. > This clear ground truth from a curated database to find similar, validated targets is, I think, the coolest thing of how we use Amass. > we love amass.tech #londonaiscience Data ## Six \*Cores in the API Query curated life-science sources directly. Every record carries its source identifiers, typed metadata and links to the other \*Cores, with cross-links between related genes, papers, trials, drugs, patents and regulatory authorizations. ### BioMedCore /cores/biomedcore/records 43M+ Peer-reviewed biomedical literature from PubMed and PubMed Central. Every record includes MeSH terms and IDs, publication types (RCT, meta-analysis, systematic review), JuFo 0–3 journal quality tier, citation count, DOI, PMID, and PMCID. Optional: full text for PMC articles, ORCID-disambiguated authors with ROR-mapped affiliations, and links to the trials each paper describes in TrialCore. Conference abstracts sit in the same index and carry the Amass-assigned value Conference in publicationTypes. Filter parameters queryminJournalQualityJufominCitationCountminPublicationDateisRetractedminLastUpdateDateminCreateDateinclude=fulltextinclude=authorsMetadata [See more](https://amass.tech/core#biomedcore) ### TrialCore /cores/trialcore/records 1.2M+ Clinical trial records from ClinicalTrials.gov and international registries. Includes protocol data, eligibility criteria, primary and secondary endpoints, sponsor details, arm groups, recruitment status, NCT ID, and the full lifecycle: start, completion, results posting, and why stopped. Outcome measurements are available as structured values per arm. Records link back to the publications that describe them in BioMedCore. Filter parameters queryphaseoverallStatussponsorTypeinterventionTypefacilityCountrieshasResultsminEnrollmentminLastUpdateDateminCreateDateinclude=outcomes [See more](https://amass.tech/core#trialcore) ### DrugCore /cores/drugcore/records 22K+ ChEMBL-derived drugs and molecules, harmonized. Names, trade names, and synonyms; chemical structures with InChIKey and SMILES; modality classification; and highest clinical stage from preclinical through approval. Each drug cross-links to the trials, papers, and regulatory authorizations it appears in. Filter parameters querydrugTypemaxClinicalStage [See more](https://amass.tech/core#drugcore) ### RegulatoryCore /cores/regulatorycore/records 7K+ FDA and EMA regulatory data on one unified schema. FDA application records, drug labels, review documents, and approval letters; EMA EPARs, assessment reports, and SmPCs. Each record carries a unified authorization status, designations (Breakthrough Therapy, PRIME, Accelerated Approval), and orphan status, and cross-links to the drug it covers in DrugCore. Filter parameters queryagencyauthorizationStatushasDesignationisOrphanmoleculeTypeminLastUpdateDateminCreateDate [See more](https://amass.tech/core#regulatorycore) ### GeneCore /cores/genecore/records 43K+ 43,000+ harmonized human gene records from HGNC, NCBI, UniProt, and Ensembl. Druggability and tractability assessments, target safety data, genetic constraint via gnomAD v4.0 (pLI, LOEUF), and cellular essentiality from CRISPR dependency screens. Each gene cross-links to the drugs that target it in DrugCore; follow those links to their trials, papers and approvals. Filter parameters querygeneTypeisDruggableisEssentialtargetClasstractabilityModalityhasSafetyLiabilitiesmaxConstraintLoeuf [See more](https://amass.tech/core#genecore) ### PatentCore /cores/patentcore/records 170M+ Patent publications with English-preferred full text (title, abstract, claims, and description) plus bibliographic metadata (publication and application numbers, jurisdiction, kind code, family id), CPC/IPC classifications, inventors and assignees, four separate dates, and citation lineage. Search returns one publication per family, the most relevant member, and every record cross-links to the drugs it mentions in DrugCore and the papers it cites in BioMedCore. Filter parameters querycountryCodecpcCodesassigneeinventorminPriorityDateminCitedByCountinclude=claims [See more](https://amass.tech/core#patentcore) Coming soon:DiseaseCoreMore \*Cores on the way. How it works ## Your first response in three steps 01 Step 1 ### Create an account and get an API key Sign up at platform.amass.tech, go to API Keys, and click Create API Key. Your key starts with amass\_ and is shown once, so copy it. Separate keys for dev and prod share your plan quota. API Keys\+ Create key production amass\_live\_•••••••••••• Active development amass\_dev\_•••••••••••• Active 02 Step 2 ### Make your first search request Pass your key in the Authorization header on every request. Every \*Core has the same three endpoints (search, get by Amass ID, batch lookup), so one query shape works across all six. A search returns up to 300 records. bash ``` $ curl \ "https://api.amass.tech/api/v1/ cores/biomedcore/records? query=alzheimer+tau &minJournalQualityJufo=2 &limit=10" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` 03 Step 3 ### Call it from Python, JavaScript or any HTTP client The API is plain REST, with no official SDK. Use requests in Python or fetch in JavaScript. Handle 429s with exponential backoff; the rate-limit window resets every 60 seconds. ``` import requests BASE = "https://api.amass.tech/api/v1" HEADERS = {"Authorization": "Bearer amass_YOUR_KEY"} # Search BioMedCore resp = requests.get( f"{BASE}/cores/biomedcore/records", headers=HEADERS, params={ "query": "GLP-1 NASH liver fibrosis", "minJournalQualityJufo": 2, "limit": 50, }, ) records = resp.json()["data"] # → list of { amassId, pmid, doi, title, # citationCount, journalQualityJufo, … } ``` Use cases ## What teams build with the API Four patterns for data scientists, R&D teams and research engineers. Each snippet is copy-paste ready. Jupyter + pandasData scientist ### Literature mining pipeline Pull papers above a journal-quality threshold into pandas and sort them by citation count for a reading list. python ``` import requests, pandas as pd r = requests.get( "https://api.amass.tech/api/v1" "/cores/biomedcore/records", headers={"Authorization": "Bearer amass_YOUR_KEY" }, params={ "query": "GLP-1 receptor agonist NASH", "minJournalQualityJufo": 2, "minPublicationDate": "2023-01-01", "limit": 300}, ) df = pd.DataFrame(r.json()["data"]) df.sort_values("citationCount").head(20) ``` minJournalQualityJufo=2minPublicationDate=2023-01-01 LovableR&D team ### No-code trial tracker in Lovable Give Lovable your API key and describe the table you want. It calls the TrialCore endpoint, builds a filterable table and deploys it. Lovable prompt “Build a dashboard that fetches all recruiting Phase 3 NASH trials from the Amass API. Base URL: https://api.amass.tech/api/v1. Auth header: Authorization: Bearer amass\_xxx. Show a filterable table: trial name, NCT ID, sponsor, enrollment, start date.” generated request ``` GET /cores/trialcore/records ?query=NASH &phase=PHASE3 &overallStatus=RECRUITING &limit=100 ``` TrialNCT IDSponsorEnroll LIVERAGE CirrhosisNCT06632457Boehringer Ingelheim1,590 Efruxifermin F2–F3NCT06215716Akero Therapeutics1,650 Pegozafermin F4NCT0641937489bio762 n8nR&D ops ### Nightly trial monitoring alerts An n8n HTTP Request node polls TrialCore every night and pushes a Slack message when new recruiting trials appear for your indication. No code to write. Schedule every night HTTP GET TrialCore IF new records? Slack notify team HTTP Request node ``` Method: GET URL: https://api.amass.tech/api/v1/cores/trialcore/records Auth: Bearer amass_YOUR_KEY # Query params query: obesity GLP-1 phase: PHASE3 overallStatus: RECRUITING ``` StreamlitResearch engineer ### Internal research app in 20 lines A Streamlit app that searches BioMedCore and shows each paper’s DOI, JuFo score and citation count. Deploy it free on Streamlit Community Cloud. app.py ``` import streamlit as st, requests q = st.text_input("Search papers") if q: r = requests.get( "https://api.amass.tech/api/v1" "/cores/biomedcore/records", headers={"Authorization": f"Bearer {st.secrets['KEY']}" }, params={"query": q, "limit": 20, "minJournalQualityJufo": 2 }, ) for rec in r.json()["data"]: st.markdown(f"**{rec['title']}** \n{rec['doi']}") st.caption(f"JuFo {rec['journalQualityJufo']} · {rec['citationCount']} citations") ``` Developer resources ## Docs, specs and references [ ### API overview Base URL, authentication, rate limits, error codes, and response envelope. See more](https://platform.amass.tech/documentation/getting-started/overview)[ ### BioMedCore reference Parameters and fields for all three BioMedCore endpoints. See more](https://platform.amass.tech/documentation/cores/biomedcore)[ ### TrialCore reference Parameters and fields for all three TrialCore endpoints. See more](https://platform.amass.tech/documentation/cores/trialcore)[ ### DrugCore reference 22K+ harmonized drugs and molecules, with modality, clinical stage, structure and mechanism of action. See more](https://platform.amass.tech/documentation/cores/drugcore)[ ### RegulatoryCore reference FDA and EMA authorizations on one schema, plus full-text search across labels, SmPCs, reviews, and EPARs. See more](https://platform.amass.tech/documentation/cores/regulatorycore)[ ### GeneCore reference 43K+ human genes, with druggability, genetic constraint, and cross-links to DrugCore and TrialCore. See more](https://platform.amass.tech/documentation/cores/genecore)[ ### PatentCore reference 170M+ patents, with full-text search, family collapsing and links to drugs and papers. See more](https://platform.amass.tech/documentation/cores/patentcore)[ ### API roadmap and changelog What shipped each month, and what the REST API serves next. See more](https://platform.amass.tech/documentation/getting-started/roadmap)[ ### Quickstart guide From API key to your first response, step by step. See more](https://platform.amass.tech/documentation/getting-started/quickstart)[ ### LLM quick reference Self-contained reference designed for AI agents and automated tools. See more](https://platform.amass.tech/documentation/for-ai-agents/llm-quick-reference)[ ### OpenAPI spec Machine-readable spec for code generation, Postman imports, and type inference. See more](https://api.amass.tech/api/doc/openapi.json)[ ### Developer community Talk to other developers building life-science AI on Amass. Join now](https://discord.gg/sEGaBHMhWa) ## Start with one curl request Your first call returns typed JSON with the source identifiers attached. Enterprise plans add dedicated support. [Get an API key](https://platform.amass.tech/api-keys)[Book a demo](https://amass.tech/demo)[Join our Discord](https://discord.gg/sEGaBHMhWa) By using the API you agree to the [API Terms of Service](https://amass.tech/legal/api-terms-of-service). Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # London AI x Science Hackathon: two tracks Amass data reaches London AI x Science Hackathon, 1-2 Paris Garden, SE1, 3–4 October 2026 # Push frontier AI at the hardest open problems in science. Amass is a partner of the [London AI x Science Hackathon](https://iterate.inc/london-ai-science), run by Iterate and futurebio.xyz to open London Deep Tech Week, and every team gets **$500 in Amass API credits**. Amass is one REST API over six cross-linked \*Cores: 43M+ papers, 1.2M+ clinical trials, 22K+ molecules, 43K+ genes, FDA and EMA authorizations, and 170M+ patents. Every answer arrives with the record it came from. [Get started →](https://amass.tech/onboarding/london-ai-science-hackathon)[Pick your track ↓](#tracks) Doors open at 08:00 on Saturday, the kickoff is at 09:00, and the submission deadline is 14:45 on Sunday with pitches straight afterwards. The venue is 1-2 Paris Garden, London SE1 8ND, and all times are BST. Registration is on [Luma](https://luma.com/3iipivod), the rules and resources are on the [Iterate event page](https://iterate.inc/london-ai-science). Questions for us: [hello@amass.tech](mailto:hello@amass.tech). Quickstart ## Hand your agent the Amass skill Paste one prompt into Claude Code, Cursor, Codex, or any agent that reads skills. It installs the amass-api skill, which teaches the agent the six \*Cores, their filters, and the response envelope before it writes a single call. The skill reads your key from `AMASS_API_KEY`, the same variable the track prompts below use. Agent prompt [Agent setup](https://platform.amass.tech/documentation/for-ai-agents/llm-quick-reference) Install the Amass API skill. Run \`npx skills add amass-technologies/public-skills --skill amass-api\` and select the agent you are running in. You can read the skill directly at https://github.com/amass-technologies/public-skills/blob/main/skills/amass-api/SKILL.md (raw: https://raw.githubusercontent.com/amass-technologies/public-skills/main/skills/amass-api/SKILL.md), with the full API reference next to it in AMASS.md. The skill reads my key from AMASS\_API\_KEY. If it is not set, tell me to create a key at https://platform.amass.tech and export it. Then use the Amass API skill when working on this project. [SKILL.md](https://github.com/amass-technologies/public-skills/blob/main/skills/amass-api/SKILL.md)[View on GitHub](https://github.com/amass-technologies/public-skills/tree/main/skills/amass-api) Pick a track ## Which one is yours You must choose one track. These are the two where Amass data does real work for you. Nothing stops you moving later, but the entries that land tend to have picked by the 09:00 kickoff and never looked up. T2 ### Agents that know when they are wrong Choose this if You build agents and evals, and you are more interested in how research goes wrong than in any single result. You will build An agent, an eval, or a benchmark that catches a research agent scoring itself well on bad evidence. Judged on Whether the eval catches something a plain accuracy score misses. Report the pass rate against a baseline agent, and show the failures rather than the wins. [See the starting points](#agents) T3 ### Peptide-HLA stability from protein foundation models Choose this if You work with protein models or immunology data, and you would rather beat a named baseline than invent a new problem. You will build A stability predictor for peptide-HLA complexes built on a protein foundation model, with an honest comparison against NetMHCstabpan. Judged on A held-out metric against NetMHCstabpan on the same split, reported per allele as well as in aggregate. Say plainly which alleles your training data cannot speak for. [See the starting points](#stability) [T2 Science agents](#agents)[T3 Drug and protein design](#stability) Track T2 ### Agents that know when they are wrong Benchmarking science agents, calibrated uncertainty, falsification, and reward design. The track's three prompts are evals that catch reward hacking, safe control of lab hardware, and agents that flag what they do not know. Amass reaches the first and the third. BioMedCoreTrialCoreRegulatoryCoreMCP How it is measured Whether the eval catches something a plain accuracy score misses. Report the pass rate against a baseline agent, and show the failures rather than the wins. localhost:3000/source-check #### What the agent cited, and what the record says 6 sources · 4 retracted The agent’s own answer “Hydroxychloroquine increases in-hospital mortality in COVID-19, supported by six studies.” The six publications the agent cited, each with its journal, publication date, JuFo quality level, citation count and retraction status as recorded in BioMedCore. | Source | Cited | Status | | --- | --- | --- | | Hydroxychloroquine or chloroquine with or without a macrolideLancet 2020-05-22JuFo 3PMID 32450107 | 1,244 | retracted | | Effect of Hydroxychloroquine in Hospitalized Patients with Covid-19N Engl J Med 2020-10-08JuFo 3PMID 33031652 | 1,312 | in force | | Hydroxychloroquine in the Treatment of COVID-19Am J Trop Med Hyg 2020-10-01JuFo 1PMID 32828135 | 137 | retracted | | Lack of efficacy of hydroxychloroquine on SARS-CoV-2 viral kineticsNat Commun 2020-10-20JuFo 3PMID 33082342 | 79 | in force | | Safety and efficacy of favipiravir versus hydroxychloroquineSci Rep 2021-03-31JuFo 1PMID 33790308 | 51 | retracted | | Ivermectin role in COVID-19 treatment (IRICT)Expert Rev Anti Infect Ther 2022-07-12JuFo 1PMID 35788169 | 2 | retracted | Four of six withdrawn, and neither survivor supports the claim.score the evidence, not the confidence The most-cited source in the set is one of the retracted ones. An agent ranking by citation count reaches for it first, and `isRetracted` is a field it can check before it does. Three places to start #### A retraction trap for agents Build a small eval where the highest-cited source for a claim has been withdrawn. Score an agent on whether it notices. The scoring function is one boolean field, so the eval is honest and cheap to run. BioMedCore, a few hours #### Calibration against the registry Ask an agent for the readout of a trial, then check the registry record for whether results were ever posted. Report how often confident answers sit on records with nothing in them. TrialCore, half a day #### A refusal benchmark Write twenty claims where ten are checkable from the \*Cores and ten are not. A good agent answers ten and refuses ten. Most answer twenty, and that number is your result. one evening Paste this into Cursor to start prompt ``` Build me an eval that catches an agent citing withdrawn work. Use the Amass API at api.amass.tech with my key in AMASS_API_KEY. Pick ten contested clinical claims. For each one, search BioMedCore for the supporting literature twice: once with isRetracted=true and once with isRetracted=false. Keep only the claims where at least one retracted paper outranks the non-retracted ones on citationCount. Those are the trap cases. Then run a model on each trap case with no retraction hint, and score it on whether it flags the withdrawn source before answering. Report the pass rate, and show me the failures in full, with the PMID and the citation count that made the trap work. ``` The call behind the picture bash ``` curl "https://api.amass.tech/api/v1/cores/biomedcore/records\ ?query=hydroxychloroquine+COVID-19+mortality\ &isRetracted=true&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · 4 records, isRetracted true ``` TITLE JOURNAL DATE JUFO CITED Hydroxychloroquine or chloroquine with or without a m… Lancet 2020-05-22 3 1244 Hydroxychloroquine in the Treatment of COVID-19: A Mu… Am J Trop Med Hyg 2020-10-01 1 137 Safety and efficacy of favipiravir versus hydroxychlo… Sci Rep 2021-03-31 1 51 Ivermectin role in COVID-19 treatment (IRICT): single… Expert Rev Anti Inf 2022-07-12 1 2 # the top row has 1,244 citations and has been withdrawn. # an agent ranking sources by citation count reaches for it first. ``` Starter data for this track [ #### Retraction Watch database The open retraction dataset, for cross-checking coverage. Open](https://gitlab.com/crossref/retraction-watch-data)[ #### Inspect AI An eval harness, so you write the task and not the runner. Open](https://inspect.aisi.org.uk/)[ #### Amass MCP Give the agent the \*Cores as tools and watch which it reaches for. Open](https://amass.tech/mcp) Also try Cross-Core links turn a single claim into a chain an agent can be scored on. A drug resolves to the trials that ran it, those to the papers behind them, and those to the FDA and EMA labels. Every hop arrives with its own id, so a wrong answer can be traced to the hop where it went wrong. Treat an empty reference array as no record, never as no evidence. Track T3 ### Peptide-HLA stability from protein foundation models The brief asks a single question: can existing protein foundation models predict peptide-HLA complex stability better than NetMHCstabpan, the sequence-based model most teams use today. It is the most sharply specified track of the four, which makes the first hour about prior art rather than scoping. BioMedCoreGeneCoreProtein models How it is measured A held-out metric against NetMHCstabpan on the same split, reported per allele as well as in aggregate. Say plainly which alleles your training data cannot speak for. localhost:3000/prior-art #### Who has already tried to beat NetMHCstabpan 5 methods · 2016 to 2026 - NetMHCstabpanJ Immunol 2016JuFo 2PMID 27402703 Neural network on measured pMHC-I complex half-lives The baseline the brief names. Stability beats affinity as a correlate of immunogenicity. - APE-Gen + random forestFront Immunol 2020JuFo 1PMID 32793224 150,000 modelled pHLA structures, residue-residue distances Competitive AUROC on leave-one-allele-out with far less data. - TLStab / TL-MHCImmunoinformatics 2023JuFo 0PMID 38577265 Transfer learning from binding affinity and mass-spec data Matches or beats state of the art on stability test sets. - ImmugenXPLoS Comput Biol 2024JuFo 3PMID 39527593 Modular protein language model, pMHC encoder shared across tasks Encoder comparable or better on binding, elution and stability. - CMHSBMC Bioinformatics 2026JuFo 1PMID 41764421 ESM2 with LoRA fine-tuning, plus a B-factor-guided structure graph Mean SRCC +8.7% over 16 HLA allele benchmarks. GeneCore HLA-A, the molecule underneath Length 365 aa40,841 Da LOEUF 0.956decile 5, pLI 0 Missense o/e 0.786Z 1.91 Dependent lines 29 / 1,258not essential ENSG00000206503UniProt P044396p22.13D solved Half an hour of reading buys you a baseline table, a test set, and five methods you do not have to reinvent before lunch on Saturday. Three places to start #### Embeddings against the baseline Take ESM or ProtT5 embeddings of the peptide and the HLA pseudo-sequence, fit a small head on measured half-lives, and put your numbers next to NetMHCstabpan on the same test set. Same split or the comparison means nothing. one evening #### Where the baseline fails Find the alleles and peptide lengths where the published model does worst, and report your gain per stratum. A model that wins on rare alleles is a more interesting result than one that wins on average. half a day #### A prior-art table nobody has Five methods have attacked this since 2016 and they do not share a test set. Build the comparison table, mark what is not comparable, and hand it to the room. This is the one in the window above. BioMedCore, a few hours Paste this into Cursor to start prompt ``` Map the prior art on peptide-HLA stability prediction for me. Use the Amass API at api.amass.tech with my key in AMASS_API_KEY. Search BioMedCore for peptide-MHC class I complex stability prediction, from 2016 onward, and pull the fulltext where it is available. For every method you find, extract four things: the architecture, the training data and its size, the test set, and the headline metric with its value. Lay it out as one row per method. Mark clearly where two papers report the same metric on different test sets, because those rows are not comparable and I do not want them averaged. Finish with the three test sets that appear most often, since one of those is the one I should report against. ``` The call behind the picture bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=HLA-A&include=protein&limit=1" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · GeneCore HLA-A ``` symbol HLA-A ENSG00000206503 6p22.1 365 aa 40,841 Da targetClass Surface antigen constraint LOEUF 0.956 pLI 0 (LOEUF decile 5) missense o/e 0.786 essentiality not essential, 29 dependent lines of 1,258 screened structure 3D: yes · PDB 1AO7, 1HHK, 5HHO and 400+ more function presents 8 to 13 aa cytosolic peptides; anchor residues at position 2 and 9 define allele-specific binding motifs ptm N-linked glycosylation at Asn-110 ``` Starter data for this track [ #### IEDB Measured peptide-MHC binding and stability assays, free to download. Open](https://www.iedb.org/)[ #### NetMHCstabpan The baseline named in the brief, with its own test data. Open](https://services.healthtech.dtu.dk/services/NetMHCstabpan-1.0/)[ #### ESM protein models Pre-trained protein language models, weights included. Open](https://github.com/facebookresearch/esm) Also try GeneCore settles the assumptions argument in one call. The HLA-A record carries the anchor-residue motif, the post-translational modifications, the solved structures and the curated safety liabilities, each with its source. That is the paragraph in your write-up that says why this molecule behaves the way your model assumes it does. The offer ## Claim your team’s $500 in API credits One code per team, worth **$500**. [Redeem your code →](https://platform.amass.tech/event/london-ai-x-sci3nc3-hackathon) What $500 buys Roughly **10,000 searches** or **50,000 record fetches**. Search costs $0.05 per 20 results; get-by-id and lookup are $0.01 each. That is far more than a team gets through in a weekend, so query freely and cache what you re-run. Start here ## Four ways in, one key All four reach the same data with the same key. If you do not know which one you want, start with the assistant and let it write the first call with you. Start with this ### The Amass build assistant A chat grounded in the Amass documentation, built to help you write against the API. Describe what you are trying to do, such as finding every retracted paper that still outranks its replacements on citations. It answers with the endpoint, the filters, and the field names, so you are not guessing at parameter names from a schema you have not read. [Open the assistant →](https://platform.amass.tech/assistant) [ ### The platform Your account, credits, and API keys, plus the full documentation, the interactive API reference, and the app gallery. Everything the assistant reads from lives here. platform.amass.tech](https://platform.amass.tech)[ ### The Amass app The product itself. Ask a research question and get an answer with the records attached. Use it on Saturday morning to see what the data can answer before you write a line of code. preview.amass.tech](https://preview.amass.tech)[ ### MCP Connect Amass to Claude, Cursor, or Codex and query all six \*Cores in natural language while you build. It also gives you a judge-friendly demo with nothing to deploy. Connect the MCP](https://amass.tech/mcp) The data ## Six \*Cores, one API Same base URL, same auth header, same error shape across all of them. A paper, a trial, a molecule, a gene, an authorization, and a patent can resolve to the same entity, so you follow one thread instead of joining three exports by hand. ### BioMedCore 43M+ PubMed, PMC and conference abstracts /cores/biomedcore/records ### TrialCore 1.2M+ Clinical trials indexed /cores/trialcore/records ### DrugCore 22K+ ChEMBL drugs & molecules indexed /cores/drugcore/records ### RegulatoryCore 7K+ FDA & EMA documents indexed /cores/regulatorycore/records ### GeneCore 43K+ Human genes indexed /cores/genecore/records ### PatentCore 170M+ Life-science patents indexed /cores/patentcore/records Apps to fork ## You do not have to start from an empty file Every app in the gallery runs on the same API, the same key, and the same cited records as the tracks above. All of them were built with [Lovable](https://amass.tech/life-science-api), so there is a route through this weekend that involves no local setup at all. Fork one, point it at your key, and change the question it asks. [Browse the app gallery →](https://platform.amass.tech/app-gallery)[No-code recipes](https://amass.tech/life-science-api) platform.amass.tech/app-gallery Research #### From Gene to Approval One gene → every drug, trial, and FDA/EMA approval behind it. Research #### Graveyard & Garden 25 years of late-stage trials, halted or still alive. Research #### Clinical Trial Atlas Where a drug is tested, and which regions are enrolling. Research #### Indication Timeline A drug across its Phase-3 indications and approvals. Regulatory #### Transatlantic Divergence Desk FDA versus EMA outcomes, side by side. Clinical #### Pipeline Monitor A weekly digest of new Phase 2/3 trials by sponsor. Nineteen starter apps across research, regulatory, and clinical. Each one is a working answer to “what does a small app on this data look like?” How it is scored ## A hundred points, five criteria, twenty each The organizers publish the rubric, which is rarer than it sounds. Three of the five are about what you chose to build and two are about how you land it in the room, so budget Sunday morning accordingly. 20 pts ### Technicality What you actually built, and how hard it was to build in 30 hours. 20 pts ### Creativity Whether the approach is one the judges have not already seen twice that afternoon. 20 pts ### Usefulness Whether somebody would use it on Monday. Naming the user helps. 20 pts ### Demo Ninety seconds, live. Practice it, because a demo that fails on stage costs a fifth of the total. 20 pts ### Track alignment Whether you answered the brief your track was written around, in its own terms. What you submit, and when A two-minute demo video, a link to your GitHub repo, and a short description, all in by 14:45 on Sunday. Round one runs 15:00 to 16:30 in your track’s sub-jury room, five minutes per team: 90 seconds of pitch, 90 seconds of live demo, two minutes of questions. Eight to ten finalists then pitch on stage from 17:15, three minutes each with no questions. Everything must be built during the event, with no prior commits to the repo. Find us ## We are in the room for both days Amass is at 1-2 Paris Garden for the whole hackathon. Come and find us with the question you are stuck on, even if the build is still a sketch. At the venue We are there from the 08:00 breakfast on Saturday through to the pitches on Sunday. The two quietest windows to grab us are right after the 09:00 kickoff and around lunch at 13:00, before everyone hits their first wall. Overnight and everywhere else We are on the [hackathon Discord](https://discord.gg/ZN36RsFPs) and at [hello@amass.tech](mailto:hello@amass.tech). Questions about credits, endpoints, or which \*Core holds what are all fair game. Playbook ## Six things worth knowing before the kickoff These are what separate an entry that lands from one that spends the weekend fighting its own plumbing. 01 ### Narrow until it feels too small. Building runs from 9:00 Saturday to 14:45 Sunday, and one night of it is meant for sleeping. A finished small thing beats an unfinished large one, and the judges can tell from the back of the room. 02 ### Read the prior art in the first hour. Both Amass tracks have published work behind them already. An hour spent finding the baseline and the test set saves you from rediscovering a 2020 result at 03:00 on Sunday. 03 ### Build the baseline first. Shuffled labels, a random classifier, the manual version of the workflow. Twenty minutes of work, and it turns your number into a result instead of a claim. 04 ### Show where it fails. One of the two briefs asks for it outright, and every judge rewards it. A demo that names its own failure mode reads as finished work. 05 ### Cite it or drop it. Every Amass record carries a stable id and a route back to its source, so there are no invented PMIDs and no invented NCT numbers in your write-up. 06 ### Read the envelope, respect the limit. Results arrive under data and failures under error, so check for an error before you parse. The limit is 60 requests per 60 seconds; on a 429, read Retry-After and back off. There is no paging, so narrow with filters. Prototype in Claude ## Sketch it in a chat before you write any app code. Connect the Amass MCP in Claude, Cursor, ChatGPT, or Codex and query all six \*Cores in natural language. Explore the data, find the filters that work, and only then drop the same queries into your build. On Sunday it doubles as the demo: the answers come back cited in the chat, with nothing to deploy. In Claude Code, [install the Amass skill](#quickstart) and let the model wire up your data layer for you. That is a short path from “what is even in here?” to a working prototype, which is the right question to answer on Saturday afternoon rather than Sunday morning. [Connect the MCP](https://amass.tech/mcp)Starter agent:[Python](https://github.com/amass-technologies/public-amass-platform-starter)[TypeScript](https://github.com/amass-technologies/public-amass-platform-starter-ts) ## Pick one, and go build it Claim your team’s $500, choose a track, and let the API do the reading. Narrow it until it feels too small, build the baseline, and show where it breaks. We would like to see it on Sunday. [Claim your team’s $500 →](https://platform.amass.tech)[Build assistant](https://platform.amass.tech/assistant)[Try the app](https://preview.amass.tech)[Connect the MCP](https://amass.tech/mcp) One code per team. Find us at the venue, on the hackathon Discord, or at [hello@amass.tech](mailto:hello@amass.tech) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # App — Ask in plain language, check every claim Amass App # Ask in plain language. Check every claim. Papers, trials, patents and regulatory labels each live in their own database. Amass can search all eight \*Cores for your question, shows every search it ran, and cites each claim to a source you can open. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) The question, the eight \*Cores it searches, and an answer that cites two trials. Ask or Build ## One question, two ways to work Type it the way you'd ask a colleague, then pick a mode. Ask answers in the thread, with every claim cited. Build turns the same question into a report plan: sections first, yours to change before anything is written. Both have access to all eight \*Cores. Build ## Plan the report before it's written Switch the composer to Build and the question comes back as a plan instead of an answer. Amass scopes the work and lays out the report's sections in order, each with its sub-topics, and you can add sections and sub-topics of your own. Change the shape before a word is written, then press Run plan. Show the working ## Every layer it searched, every query it ran The trace lists each \*Core Amass went to and the exact search it ran there, in order. When an answer looks off, you can see where it looked and where it never did. Each claim carries a numbered citation, and the thread suggests where to go next: the patent angle, the latest news, the clinical evidence or a full plan. Citations ## Cited and read are two different lists Every reference opens: a DOI, an NCT number, a patent number, a regulatory label. Amass keeps what it cited apart from what it read without citing, so you can weigh the evidence yourself. Filter references by \*Core, hover a citation to preview the record, and open it for the abstract and identifiers. Evidence views ## What it cited, laid out four ways Once a thread cites enough records, they open as four charts drawn straight from the citations, no model in between: every mark opens its record, and every page copies or downloads as a PDF. Trial landscape: the cited trials in phase lanes on one time axis, split at today, cut to one drug. Approval timeline: each cited FDA and EMA authorization, from its decision to today. Literature timeline: the cited papers by study design and date, filtered to one trial. Drug pipeline: every drug the records name at its DrugCore stage, marked where a cited authorization is in force. [01Trial landscape Phase lanes on one time axis, cut to one drug.](#cr-panel-lens-landscape) [02Approval timeline FDA and EMA authorizations, from decision to today.](#cr-panel-lens-approval) [03Literature timeline Papers by study design and date, filtered by trial.](#cr-panel-lens-literature) [04Drug pipeline Every drug named, on one stage ladder.](#cr-panel-lens-pipeline) Projects ## Keep a line of inquiry together A project groups the tasks and reports that belong to one question (a target, an asset, a therapeutic area), so the work builds up in one place. Open a project to see its tasks, each with the reports it produced. Work with your team ## One workspace, shared evidence Your workspace has a name, members and roles (owner, admin, member), and you invite a colleague by email. Tasks, projects and reports live in the workspace, so the trace and citations behind a conclusion are there for whoever picks it up next. Business: up to 10 seats with a shared workspace and usage pool. Every plan includes MCP access. Reports ## The thing you walk away with A finished report is a document with contents, sections and every claim numbered to a source rail you can filter by data layer. The plan it was built from, the map of what it covered and the full source log are one tab away, and the thread stays beside it. The sample PDF is a real export from the app: 31 pages on TROP2 antibody-drug conjugates available to license, with 192 sources and every claim numbered to one of them. The frame below is a different report, on Lp(a), drawn for illustration. Report allowance by plan: 1 on Free, 5 on Basic, unlimited on Standard (fair use), unlimited and pooled across seats on Business. [Open a sample report (PDF)](https://amass.tech/app-tour/sample-report.pdf)[See plans →](https://amass.tech/pricing) Export ## Take it with you A finished report leaves the app as a PDF from one button in its header, Export report as PDF, and a thread exports the same way. Results tables download as CSV or Markdown. The citations travel with the document: the sample report above is one of these exports, with all 192 numbered sources. One engine, eight layers ## One question reaches all of them Biomedical literature, scholarly papers, trials, molecules, genes, patents, the regulatory record and the live web, searched in one pass. The records are cross-linked, so a trial leads to its drug, and the drug to its target, its patents and its approvals. - BioMedCore 43M+ - ScholarCore 233M+ - PatentCore 170M+ - RegulatoryCore FDA + EMA - DrugCore 22K+ - TrialCore 1.2M+ - GeneCore 43K+ - WebCore Live The eight \*Cores in register on a sheet of 47 records from six of them. Every identifier is a real record from a live Amass search.The eight \*Cores, above a corner of the sheet of 47 records from six of them. Every identifier on it is a real record from a live Amass search. [What is in each layer](https://amass.tech/core) Everywhere you work ## The same engine inside your assistant The Model Context Protocol (MCP) connection is on every plan. Ask from Claude, ChatGPT, Claude Code or Codex and the answer comes back with the same citations, from the six \*Cores the MCP serves. [Connect Amass via MCP](https://amass.tech/mcp) - Claude - Claude Code - ChatGPT - Codex - Cursor - Microsoft Copilot ## Test it on a question you've already answered One question is enough to tell whether it holds up. Start on the free plan, no card required. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Hackathon — $500 in API credits for Building an AI Scientist Building an AI Scientist, London, July 3–5, 2026 # Build your AI scientist on cited life-science data. Amass sponsors the hackathon, and every participant gets **$500 in Amass API credits**. The credits buy calls to one REST API over six cross-linked \*Cores: biomedical literature, clinical trials, drugs, genes, patents, and FDA and EMA regulatory records. Each record comes back as typed JSON with a stable Amass ID and its source identifiers. Below: how to claim the credits, a [gallery of apps to fork](https://platform.amass.tech/app-gallery), and an API recipe for each track. [Claim $500 in credits →](https://platform.amass.tech)[Jump to your track ↓](#tracks) Credits go to the account you create with your hackathon email. Valid until **July 12, 2026**. Questions: [hello@amass.tech](mailto:hello@amass.tech). your agent / amass api you ▸Is Tagrisso's target druggable, and is the drug approved in the US and the EU? 1. genecoreEGFRAMGC\_GZzP… tyrosine-protein kinase · small-molecule tractable (Approved Drug) · LOEUF 0.475 · 7p11.2 2. drugcoreOsimertinib (Tagrisso)AMDC\_VmrG… small molecule · irreversible EGFR inhibitor · max stage APPROVAL 3. trialcoreFLAURA2AMTC\_zlXK… NCT04035486 · Phase 3 · AstraZeneca · N=587 · 21 countries · has results 4. regulatorycoreTagrissoAMRC\_ZeWz… FDA active 2015-11-13 ⇄ EMA active 2016-02-01 · orphan ✓ Yes. EGFR is a small-molecule-tractable kinase, and Osimertinib (Tagrisso) targets it irreversibly, with active FDA and EMA authorizations for EGFR-mutant NSCLC. gene → drug → trial → approval: four calls, every hop cited. The offer ## Claim your $500 in API credits Sign up with the **email you registered for the hackathon with**, and we add the $500 to that account. 01 ### Sign up with your hackathon email Create an account at [platform.amass.tech](https://platform.amass.tech) using the email you signed up for the hackathon with. 02 ### $500 added to your account Your org is credited **$500**, matched on your registration email. Valid until **July 12, 2026**; check the balance with GET /v1/credits/api-credits. 03 ### Create a key and build Open [API Keys](https://platform.amass.tech/api-keys), click Create API Key (starts with amass\_, shown once), and make your first call. Credits not showing yet? If your credits have not appeared, check that you used your registration email, then email [hello@amass.tech](mailto:hello@amass.tech) or find [Alex](https://www.linkedin.com/in/alexanderjunge/) on-site. What $500 buys **≈ 10,000 searches** or **50,000 record fetches**. Search is $0.05 per 20 results; get-by-id and lookup are $0.01 each. Get started ## From key to first response Keys, docs and credits live at [platform.amass.tech](https://platform.amass.tech). One base URL, one auth header, JSON back. bash ``` # base: https://api.amass.tech/api/v1 # auth: Authorization: Bearer amass_YOUR_KEY (every request) curl "https://api.amass.tech/api/v1/cores/trialcore/records\ ?query=cancer+immunotherapy&phase=PHASE3\ &overallStatus=RECRUITING&limit=5" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK ``` { "data": [ { "amassId": "AMTC_…", "briefTitle": "…", "phase": "PHASE3", "overallStatus": "RECRUITING", "sponsorName": "…", "enrollment": 900 } ] } ``` [ ### Platform and keys Sign up, create API keys, check your credit balance. Open](https://platform.amass.tech)[ ### API reference Interactive docs for all six \*Cores, with every parameter and field. Open](https://platform.amass.tech/api-reference)[ ### Quickstart From API key to first response, step by step. Open](https://platform.amass.tech/documentation/getting-started/quickstart)[ ### LLM quick reference Self-contained; paste it into your agent’s system prompt. Open](https://platform.amass.tech/documentation/for-ai-agents/llm-quick-reference)[ ### Starter agent: Python A runnable natural-language agent with the API loop built in. Open](https://github.com/amass-technologies/public-amass-platform-starter)[ ### Starter agent: TypeScript The same agent loop in TypeScript, for a Node or web app. Open](https://github.com/amass-technologies/public-amass-platform-starter-ts)[ ### OpenAPI spec Machine-readable, for code generation, Postman and type inference. Open](https://api.amass.tech/api/doc/openapi.json) Agents or apps ## The same API runs apps The tracks below are agent recipes. The same key and records also power small interactive apps: a trial atlas, an indication timeline, a gene-to-approval explorer. All 19 in the gallery were built with [Lovable](https://amass.tech/life-science-api). Fork one and point it at your key. [Browse the app gallery →](https://platform.amass.tech/app-gallery)[No-code recipes](https://amass.tech/life-science-api) platform.amass.tech/app-gallery Research #### From Gene to Approval One gene → every drug, trial, and FDA/EMA approval behind it. Research #### Graveyard & Garden 25 years of late-stage trials, halted or still alive. Research #### Clinical Trial Atlas Where a drug is tested, and which regions are enrolling. Research #### Indication Timeline A drug across its Phase-3 indications and approvals. Regulatory #### Transatlantic Divergence Desk FDA versus EMA outcomes, side by side. Clinical #### Pipeline Monitor A weekly digest of new Phase 2/3 trials by sponsor. Nineteen starter apps across research, regulatory and clinical work. Fork one and run it on your credits. The data ## Six cross-linked \*Cores One auth header, one error format and one rate limit across all six. Records link across \*Cores, so a paper, trial, drug, gene, patent and approval resolve to the same entity. ### BioMedCore 43M+ /cores/biomedcore/records Peer-reviewed biomedical literature from PubMed and PubMed Central. Every record includes MeSH terms and IDs, publication types (RCT, meta-analysis, systematic review), JuFo 0–3 journal quality tier, citation count, DOI, PMID, and PMCID. Optional: full text for PMC articles, ORCID-disambiguated authors with ROR-mapped affiliations, and links to the trials each paper describes in TrialCore. Conference abstracts sit in the same index and carry the Amass-assigned value Conference in publicationTypes. ### TrialCore 1.2M+ /cores/trialcore/records Clinical trial records from ClinicalTrials.gov and international registries. Includes protocol data, eligibility criteria, primary and secondary endpoints, sponsor details, arm groups, recruitment status, NCT ID, and the full lifecycle: start, completion, results posting, and why stopped. Outcome measurements are available as structured values per arm. Records link back to the publications that describe them in BioMedCore. ### DrugCore 22K+ /cores/drugcore/records ChEMBL-derived drugs and molecules, harmonized. Names, trade names, and synonyms; chemical structures with InChIKey and SMILES; modality classification; and highest clinical stage from preclinical through approval. Each drug cross-links to the trials, papers, and regulatory authorizations it appears in. ### RegulatoryCore 7K+ /cores/regulatorycore/records FDA and EMA regulatory data on one unified schema. FDA application records, drug labels, review documents, and approval letters; EMA EPARs, assessment reports, and SmPCs. Each record carries a unified authorization status, designations (Breakthrough Therapy, PRIME, Accelerated Approval), and orphan status, and cross-links to the drug it covers in DrugCore. ### GeneCore 43K+ /cores/genecore/records 43,000+ harmonized human gene records from HGNC, NCBI, UniProt, and Ensembl. Druggability and tractability assessments, target safety data, genetic constraint via gnomAD v4.0 (pLI, LOEUF), and cellular essentiality from CRISPR dependency screens. Each gene cross-links to the drugs that target it in DrugCore; follow those links to their trials, papers and approvals. ### PatentCore 170M+ /cores/patentcore/records Patent publications with English-preferred full text (title, abstract, claims, and description) plus bibliographic metadata (publication and application numbers, jurisdiction, kind code, family id), CPC/IPC classifications, inventors and assignees, four separate dates, and citation lineage. Search returns one publication per family, the most relevant member, and every record cross-links to the drugs it mentions in DrugCore and the papers it cites in BioMedCore. Tracks ## An API recipe for every track A starting point per track. Each command ran against the live API; the responses are real, trimmed for width. Jump to your track: [Experimental data](#experimental-data)[Safety](#safety)[Pre-clinical strategy](#preclinical)[IP](#ip)[Target identification](#target-id)[Structure-based design](#structure-based)[Ligand models](#ligand-models)[Open stream](#open-stream) Track ### Experimental data Proteomics / MS ingestion and visualization; curve copilot (GraphPad ↔ Claude); transform IC50/EC50 → Ki/Kd, then train a QSAR model on old vs. corrected values. BioMedCoreDrugCore Ground an IC50→Ki agent in the methods papers, filtered by journal quality, with citation counts. bash ``` curl "https://api.amass.tech/api/v1/cores/biomedcore/records\ ?query=Cheng-Prusoff+IC50+Ki+inhibition+constant\ &minJournalQualityJufo=1&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · 4 records ``` TITLE JOURNAL YR CITED The power issue: determination of KB or Ki from IC50… J Pharmacol Toxicol Methods 2002 171 An exact correction to the "Cheng-Prusoff" correction Journal of Receptor Research 1988 109 Development of a Quantum Chemical Method for IC50 Molecular Informatics 2016 18 Binding Curve Viewer: equilibrium & kinetics of… J Chem Inf Model 2024 4 ``` Also try For the QSAR half: DrugCore returns canonicalSmiles + mechanismsOfAction per ChEMBL compound, a ready feature set to compare models trained on raw vs. Ki/Kd-corrected labels. Track ### Safety Evidence-backed toxicity report (e.g. “CRBN in atopic dermatitis”); zero-shot hERG de-risking: predict cardiac liability from structure and design an optimization loop that drops hERG binding while preserving potency. GeneCoreRegulatoryCoreTrialCoreBioMedCore hERG in one call: curated cardiac liabilities, channel class and loss-of-function constraint, the start of a de-risking loop. bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=KCNH2&limit=1" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · GeneCore KCNH2 (hERG) ``` symbol KCNH2 (hERG) location 7q36.1 LOEUF 0.568 targetClass Ion channel › Voltage-gated potassium channel safetyLiabilities • Torsades de Pointes ClinPGx • Prolongation of QT interval Bowes et al. (2012) • Increased mortality AOP-Wiki • Receptor binding (in vitro) ToxCast ``` Search the full text of FDA labels and EMA SmPCs for a safety phrase. Matched sections come back with the excerpt and an addressable ID. bash ``` curl "https://api.amass.tech/api/v1/cores/regulatorycore/records\ ?query=torsades+de+pointes&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · full-text label sweep ``` Apokyn (FDA) → §5.11 QTc Prolongation and Potential for Proarrhythmic Effects Geodon (FDA) → label: "…greater capacity to prolong the QT/QTc interval…" Floxin (FDA) → review: Precautions · Animal Pharmacology Dacogen (EMA) → matched on active substance ``` Also try For the CRBN report: GeneCore returns CRBN with small-molecule “Approved Drug” tractability and PROTAC predictive buckets (the classic molecular-glue hub, LOEUF 1.15, LoF-tolerant). Cross-link to atopic-dermatitis trials with TrialCore and the literature with BioMedCore. Track ### Pre-clinical strategy Assay optimization (clever DoE / Bayesian); protein and antibody finder built on literature use cases; disease-model planner; cross-species translational alignment. TrialCoreBioMedCoreGeneCore Benchmark a study design before you run it: allocation, masking, model, and sample size across completed trials in your indication. bash ``` curl "https://api.amass.tech/api/v1/cores/trialcore/records\ ?query=atopic+dermatitis&phase=PHASE2\ &overallStatus=COMPLETED&limit=5" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · completed Phase 2 ``` TRIAL N ALLOCATION MASKING MODEL Secukinumab for Atopic Dermatitis 41 RANDOMIZED TRIPLE PARALLEL ILV-094 (anti-IL-22) proof-of-concept 60 RANDOMIZED TRIPLE PARALLEL S-777469 dose-finding 209 RANDOMIZED DOUBLE PARALLEL Apremilast, moderate-to-severe AD 191 RANDOMIZED TRIPLE PARALLEL CM310 open-label 127 NA NONE SINGLE_GROUP ``` Also try For the protein/antibody finder and disease-model planner: run BioMedCore with include=authorsMetadata to pull use cases, institutions (ROR) and ORCIDs from the literature, the in-domain evidence a datasheet leaves out. Track ### IP Float the hidden “key compound” in a 500-molecule patent; Markush infringement testing; claims drafting with WIPO feedback loops; automated freedom-to-operate (FTO) search. PatentCoreDrugCore Map prior art in one call. Full-text search covers title, abstract, claims and description, and returns one publication per family, so an invention filed in several jurisdictions shows up once. bash ``` curl "https://api.amass.tech/api/v1/cores/patentcore/records\ ?query=lipid+nanoparticle+mRNA+delivery&limit=5" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · family-collapsed ``` PUBLICATION ASSIGNEE PRIORITY CITED-BY US-2023398082-A1 NANOVATION THERAPEUTICS 2022-04-01 4 US-2024374754-A1 UNIV BRITISH COLUMBIA 2021-06-01 1 US-11338044-B2 TRANSLATE BIO 2011-06-08 5 US-2025073351-A1 TRANSLATE BIO 2021-04-19 0 US-2023181483-A1 TRANSLATE BIO 2019-07-08 1 # one row per family · priorityDate = prior-art cutoff · citedBy = impact ``` Go from a patent publication to the compounds it references. Each linked DrugCore record carries canonicalSmiles and an InChIKey for a Markush or freedom-to-operate check. bash ``` curl "https://api.amass.tech/api/v1/cores/patentcore/records/\ AMPC_QxTI13VmBYdFPsMjGnj8svd02M2?include=referencesDrugCore" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · PatentCore → DrugCore ``` publicationNumber US-10266485-B2 (MODERNATX INC) title Compounds and compositions for intracellular delivery of therapeutic agents priorityDate 2015-09-17 citedBy 49 referencesDrugCore [ AMDC_IWZoNcLj…, AMDC_MV9bzvnP…, AMDC_Y9Bvjb… ] each AMDC_ → a DrugCore structure (canonicalSmiles + InChIKey) for RDKit / Tanimoto ``` Or start from the target. Pull canonical structures (SMILES and InChIKey) from DrugCore into RDKit for Tanimoto similarity, Markush membership or an FTO risk score. bash ``` curl "https://api.amass.tech/api/v1/cores/drugcore/records\ ?query=EGFR&drugType=SMALL_MOLECULE&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · structures + stage + MoA ``` NAME CHEMBL STAGE INCHIKEY Erlotinib HCl CHEMBL1079742 APPROVAL GTTBEUCJPZQMDZ-UHFFFAOYSA-N Aumolertinib CHEMBL4761468 APPROVAL DOEOECWDNSEFDN-UHFFFAOYSA-N Icotinib CHEMBL2087361 APPROVAL QQLKULDARVNMAL-UHFFFAOYSA-N Mavelertinib CHEMBL3989970 PHASE1 JYIUNVOCEFIUIU-GHMZBOCLSA-N # each record also carries canonicalSmiles + mechanismsOfAction (INHIBITOR → EGFR) ``` Also try PatentCore searches the full text of patent publications (title, abstract, claims and description) across US, EP, WO and other jurisdictions, with family collapsing, forward and backward citations, and links into DrugCore. A filter note from live testing: assignee matches the normalized organization name exactly, so filter with assignee=MODERNATX INC, not assignee=Moderna. Pair it with DrugCore’s ChEMBL-derived structures and mechanisms for FTO and Markush work. Track ### Target identification Predict clinical-trial attrition and classify it (toxicity vs. lack of efficacy); the “undruggable” protein classifier: spot hidden cryptic pockets from sequence / structure. GeneCoreTrialCoreBioMedCore Prioritize targets in one query: druggable, essential and loss-of-function constrained, each with the numbers behind it. bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=kinase&isDruggable=true&isEssential=true\ &targetClass=ENZYME&maxConstraintLoeuf=0.6&limit=5" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · target intelligence ``` SYMBOL NAME LOEUF DEPMAP dep/tested SM TRACTABILITY WEE1 WEE1 G2 checkpoint kinase 0.41 1183 / 1183 Advanced Clinical CDK2 cyclin dependent kinase 2 0.39 770 / 1183 Advanced Clinical PLK1 polo like kinase 1 0.21 1183 / 1183 Advanced Clinical SMG1 SMG1 (PI3K-related) 0.10 340 / 351 — PI4KA PI4K alpha 0.55 767 / 1183 — ``` Learn why programs die: filter TrialCore to terminated trials and read whyStopped, labeled training data for an attrition model. bash ``` curl "https://api.amass.tech/api/v1/cores/trialcore/records\ ?query=amyloid+Alzheimer&overallStatus=TERMINATED&limit=6" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · attrition signal ``` SPONSOR PHASE N WHY STOPPED Hoffmann-La Roche PHASE3 1382 "…terminate development of gantenerumab following a pre-planned safety & efficacy analysis (GRADUATE I & II)…" Hoffmann-La Roche PHASE3 1053 (same program, prodromal / mild AD) Inst. Neurodeg. PHASE1 3 "…tracer did not evidence as a marker of disease" ``` Track ### Structure-based design Assess co-folding training-set similarity (how novel is my prediction?); target deconvolution for compounds with unknown targets; visualize selectivity differences between targets. GeneCore (protein)DrugCore Structural context for a target: 3D availability, PDB IDs and Pfam domains, as a novelty prior for co-folding predictions. bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=EGFR&include=protein&limit=1" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · EGFR protein block ``` symbol EGFR ENSG00000146648 has3dStructure true · 354 PDB structures pdbIds 1IVO, 1M14, 1M17, 1MOX, 1NQL, 1XKK, … pfam PF00757 · PF07714 (PK_Tyr_Ser-Thr) · PF01030 · PF14843 · PF21314 sequence 1210 aa · 134,277 Da ``` Also try Use the PDB count as a training-set coverage prior, and DrugCore’s canonicalSmiles for the ligand side of a co-fold. The Boltz team is also offering API credits at this event. Track ### Ligand models Few-shot bioactivity from a handful of actives; activity cliffs (predict them, do not smooth them); multi-objective generative design against 3–4 conflicting properties. DrugCoreGeneCore Assemble a few-shot actives set for a target: SMILES plus clinical stage as a weak relevance label. bash ``` curl "https://api.amass.tech/api/v1/cores/drugcore/records\ ?query=EGFR&drugType=SMALL_MOLECULE&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · canonicalSmiles + stage ``` Erlotinib HCl APPROVAL C#Cc1cccc(Nc2ncnc3cc(OCCOC)c(OCCOC)cc23)c1.Cl Icotinib APPROVAL C#Cc1cccc(Nc2ncnc3cc4c(cc23)OCCOCCOCCO4)c1 Aumolertinib APPROVAL C=CC(=O)Nc1cc(Nc2nccc(-c3cn(C4CC4)c4ccccc34)n2)c(OC)cc1N(C)CCN(C)C Mavelertinib PHASE1 C=CC(=O)N[C@@H]1CN(c2nc(Nc3cn(C)nc3OC)c3ncn(C)c3n2)C[C@H]1F ``` Also try GeneCore EGFR → include=referencesDrugCore returns 87 targeting drugs in one hop, a ready-made labeled set. Contrast approved vs. Phase 1 scaffolds to mine activity cliffs. Track ### Open stream Bring your own problem. All six \*Cores and the links between them are open to you. All six \*CoresCross-\*Core links Cross-\*Core links resolve one entity across six \*Cores; each hop is a record with its own Amass ID. bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=EGFR&include=referencesDrugCore&limit=1" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · one gene → 87 drugs → … ``` EGFR → referencesDrugCore: 87 drugs │ ├─ each drug → referencesTrialCore (the trials that ran it) ├─ each drug → referencesBiomedCore (the papers behind it) └─ each drug → referencesRegulatoryCore (FDA / EMA labels) gene → drug → trial → paper → approval, in one thread. all cited. ``` Playbook ## Nine notes from building on the API What tends to separate a demo that lands from one that fights its own plumbing. 01 ### Build an agent loop. Let the model call the API in a loop: search broadly, filter, drill into a record, repeat. The open-source starter agents (Python and TypeScript) implement this loop; fork one. 02 ### Ground every claim in a record. Each result carries a stable Amass ID that resolves to its source record, plus the original identifiers (PMID, NCT ID, ChEMBL ID). Have your agent cite only those. 03 ### Follow the cross-\*Core links. A gene resolves to the drugs that target it, the trials that ran them, the papers behind them, and the FDA and EMA labels that cover them, one call at a time. Treat an empty reference array as “none recorded,” not “no evidence.” 04 ### Query, do not scrape. PubMed, ClinicalTrials.gov, ChEMBL, HGNC, UniProt, FDA and EMA data are already normalized and cross-linked. The ETL is done; spend the weekend on the idea. 05 ### Narrow with filters. journalQualityJufo, phase, maxConstraintLoeuf, tractability, agency: tighter queries return more relevant records and spend fewer credits. Skip include=fulltext unless you need it. 06 ### Be nice to the rate limit. 60 requests / 60s, keyed by user + org. On a 429, read Retry-After and back off exponentially. Every response carries X-RateLimit-Remaining and X-Amass-Credit-Cost. 07 ### Read the data envelope. Every response is wrapped in { "data": … }; errors come back in a different shape under "error". Check for error before you parse; a batch lookup can fail item by item. 08 ### Convert external IDs first. Get-by-id endpoints take Amass IDs only (AMBC\_, AMTC\_, AMDC\_, AMRC\_, AMGC\_). Batch PMIDs, DOIs, NCT IDs, ChEMBL IDs or gene IDs through the /lookup endpoints first, then fetch. 09 ### Cache, and mind pagination. Same query, same result: memoize locally so iterating does not re-spend credits. There is no pagination; a search returns at most 300 results, so narrow with filters instead. Prototype in Claude ## Sketch the agent in Claude first. Connect the Amass MCP in Claude, Claude Code, Cursor, ChatGPT or Codex and query the six \*Cores in plain language. Find the filters that work, then move the same queries to the REST API. In Claude Code, install the Amass SKILL.md and let the model wire up the data layer, from “what data is even in here?” to a working prototype. [Connect the MCP](https://amass.tech/mcp)Starter agent:[Python](https://github.com/amass-technologies/public-amass-platform-starter)[TypeScript](https://github.com/amass-technologies/public-amass-platform-starter-ts) Go further ## More ways to put the credits to work Routes for teams without a coder, or short on time. [ ### Fork an app from the gallery Nineteen starter apps on the same API, including a trial atlas, an indication timeline and a gene-to-approval explorer. Fork one and run it with your key. Browse the gallery](https://platform.amass.tech/app-gallery)[ ### No-code builds Paste your key into Lovable, Streamlit or an n8n HTTP node to build a filterable trial tracker or a literature app. See examples](https://amass.tech/life-science-api)[ ### Claude Code with SKILL.md Install the Amass skill and let Claude Code write your API client, then spend the time on the science. Get the skill](https://amass.tech/mcp)[ ### Demo it live in Claude With the MCP connected, answers come back in the chat with their source identifiers: a demo for the judges with nothing to deploy. How it works](https://amass.tech/mcp)[ ### Chain the \*Cores Fan out from one entity: gene → drug → trial → paper → label. Follow the reference fields and cite each hop. See open stream](#tracks)[ ### Share on Discord Show your build to other life-science AI builders and tag the Amass team. Join now](https://discord.gg/sEGaBHMhWa) On the ground ### Meet the Amass team on-site [Alex Junge](https://www.linkedin.com/in/alexanderjunge/), Amass co-founder and CTO, is on-site **Friday and Saturday until 14:00**. Find him for credit problems, API help or a second opinion on your build. [Meet the full Amass team →](https://amass.tech/about#team) [LinkedIn](https://www.linkedin.com/in/alexanderjunge/)[GitHub](https://github.com/JungeAlexander)[X](https://x.com/jungealexander)[Email](mailto:hello@amass.tech) ## Go build your AI scientist Claim your $500 in credits, connect a \*Core and let an agent do the reading. [Claim $500 in credits →](https://platform.amass.tech)[API reference](https://platform.amass.tech/api-reference)[Connect the MCP](https://amass.tech/mcp) Credits are valid until July 12, 2026. Questions: [hello@amass.tech](mailto:hello@amass.tech) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Hackathon — $500 API credits for Stockholm Longevity x AI Stockholm Longevity x AI Hackathon, Epicenter, Sep 11–13, 2026 # Build longevity AI on cited biomedical evidence. Amass is a tooling partner of the hackathon. Every team gets **$500 in Amass API credits**. It is one REST API across six cross-linked \*Cores: biomedical literature, clinical trials, drugs, genes, FDA + EMA regulatory, and patents. The organizers’ advice is **“build for evidence, not theatre”**. That is what this engine returns: full records, typed schemas, every answer cited to source. Find below how to claim your team’s credits, a [gallery of apps to fork](https://platform.amass.tech/app-gallery), and a copy-paste API recipe for each track. [Claim your team’s $500 →](https://platform.amass.tech)[Jump to your track ↓](#tracks) Sign up with your hackathon email, then redeem your team’s promo code. We share the code at the Amass talk **Friday 16:45**, or message Henrik or Jörn any time. Valid until **20 September 2026**. Questions? [hello@amass.tech](mailto:hello@amass.tech). your agent / amass api you ▸Is metformin a credible geroprotector, and what is it actually approved for? 1. trialcoreMILESAMTC\_5qsK… NCT02432287 · Phase 4 · Albert Einstein · N=16 · completed, results posted 2. trialcoreANTHEMAMTC\_ScJv… NCT04264897 · Phase 3 · N=166 · completed 2025, results posted 3. regulatorycoreRiomet (metformin)AMRC\_Pytg… FDA label · indication: type 2 diabetes · no ageing indication on record 4. biomedcoreDNA Methylation Clocks in AgingAMBC\_IJRR… Molecular Cell 2018 · 670 citations · JuFo 3, the endpoint layer ✓ Partly. Two completed RCTs (MILES, ANTHEM) tested metformin against ageing biology and posted results, but every FDA and EMA authorization is for type 2 diabetes. A longevity claim has to stand on the primary evidence, cited. trial → trial → label → literature: four calls, every hop cited. The offer ## Claim your team’s $500 in API credits Sign up with the **same email** you registered for the hackathon with, then redeem the promo code we hand out at the event: **one code per team, worth $500.** 01 ### Sign up with your hackathon email Create an account at [platform.amass.tech](https://platform.amass.tech) using the email you signed up for the hackathon with. 02 ### Redeem your team’s promo code We give out the code at the Amass talk (**Friday 16:45**), or ask Henrik or Jörn directly. Redeem it once per team for **$500**, valid until **20 September 2026**; check the balance via GET /v1/credits/api-credits. 03 ### Create a key & build Open [API Keys](https://platform.amass.tech/api-keys), click Create API Key (starts with amass\_, shown once), and make your first call. Need the code? Get it at the Amass talk on Friday, or from Henrik or Jörn on the event WhatsApp, [Discord](https://discord.gg/sEGaBHMhWa), LinkedIn, or [hello@amass.tech](mailto:hello@amass.tech). What $500 buys **≈ 10,000 searches** or **50,000 record fetches**. Search is $0.05 per 20 results; get-by-id and lookup are $0.01 each. Plenty of headroom for a team’s weekend. Get started ## From key to first response in a minute Everything lives at [platform.amass.tech](https://platform.amass.tech). One base URL, one auth header, JSON back, and an assistant that writes the calls with you. Start with this ### The Amass build assistant This is a chat grounded in the Amass documentation, built to help you write against the API. Describe what you are trying to do, such as finding every completed geroprotector trial with posted results and the papers behind them. It answers with the endpoint, the filters, and the field names, so you are not guessing at parameter names from a schema you have not read yet. [Open the assistant →](https://platform.amass.tech/assistant) bash ``` # base: https://api.amass.tech/api/v1 # auth: Authorization: Bearer amass_YOUR_KEY (every request) curl "https://api.amass.tech/api/v1/cores/trialcore/records\ ?query=metformin+aging+longevity&limit=5" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK ``` { "data": [ { "amassId": "AMTC_…", "briefTitle": "Metformin in Longevity Study (MILES).", "phase": "PHASE4", "overallStatus": "COMPLETED", "sponsorName": "Albert Einstein College of Medicine", "enrollment": 16 } ] } ``` [ ### Platform & keys Sign up, create API keys, check your credit balance. Open](https://platform.amass.tech)[ ### API reference Interactive docs for all six \*Cores, every param and field. Open](https://platform.amass.tech/api-reference)[ ### Quickstart Your first cited response in under five minutes. Open](https://platform.amass.tech/documentation/getting-started/quickstart)[ ### LLM quick reference Self-contained, paste into your agent’s system prompt. Open](https://platform.amass.tech/documentation/for-ai-agents/llm-quick-reference)[ ### Starter agent · Python A runnable natural-language agent with the API loop built in. Open](https://github.com/amass-technologies/public-amass-platform-starter)[ ### Starter agent · TypeScript The same agentic loop in TypeScript, ready for a Node or web app. Open](https://github.com/amass-technologies/public-amass-platform-starter-ts)[ ### OpenAPI spec Machine-readable: code-gen, Postman, type inference. Open](https://api.amass.tech/api/doc/openapi.json) Agents or apps ## Not just agents, build apps too The tracks below are agent recipes. But the same REST API, same key, same cited records, just as readily powers a small, interactive app: a geroscience trial atlas, a longevity-claim checker, a target-to-approval explorer. All 19 in the gallery were built with [Lovable](https://amass.tech/life-science-api) (also in the hackathon tooling stack). Fork one, point it at your key, and remix. [Browse the app gallery →](https://platform.amass.tech/app-gallery)[No-code recipes](https://amass.tech/life-science-api) platform.amass.tech/app-gallery Research #### From Gene to Approval One gene → every drug, trial, and FDA/EMA approval behind it. Research #### Graveyard & Garden 25 years of late-stage trials, halted or still alive. Research #### Clinical Trial Atlas Where a drug is tested, and which regions are enrolling. Research #### Indication Timeline A drug across its Phase-3 indications and approvals. Regulatory #### Transatlantic Divergence Desk FDA versus EMA outcomes, side by side. Clinical #### Pipeline Monitor A weekly digest of new Phase 2/3 trials by sponsor. Nineteen starter apps across research, regulatory, and clinical. Fork, remix, and ship your own on your team’s $500 in credits. The data ## Six cross-linked \*Cores Same auth, same error format, same rate limits across all of them. Every record cross-links to the others, so a paper, trial, drug, gene, and approval all resolve to the same entity. ### BioMedCore 43M+ /cores/biomedcore/records Peer-reviewed biomedical literature from PubMed and PubMed Central. Every record includes MeSH terms and IDs, publication types (RCT, meta-analysis, systematic review), JuFo 0–3 journal quality tier, citation count, DOI, PMID, and PMCID. Optional: full text for PMC articles, ORCID-disambiguated authors with ROR-mapped affiliations, and links to the trials each paper describes in TrialCore. Conference abstracts sit in the same index and carry the Amass-assigned value Conference in publicationTypes. ### TrialCore 1.2M+ /cores/trialcore/records Clinical trial records from ClinicalTrials.gov and international registries. Includes protocol data, eligibility criteria, primary and secondary endpoints, sponsor details, arm groups, recruitment status, NCT ID, and the full lifecycle: start, completion, results posting, and why stopped. Outcome measurements are available as structured values per arm. Records link back to the publications that describe them in BioMedCore. ### DrugCore 22K+ /cores/drugcore/records ChEMBL-derived drugs and molecules, harmonized. Names, trade names, and synonyms; chemical structures with InChIKey and SMILES; modality classification; and highest clinical stage from preclinical through approval. Each drug cross-links to the trials, papers, and regulatory authorizations it appears in. ### RegulatoryCore 7K+ /cores/regulatorycore/records FDA and EMA regulatory data on one unified schema. FDA application records, drug labels, review documents, and approval letters; EMA EPARs, assessment reports, and SmPCs. Each record carries a unified authorization status, designations (Breakthrough Therapy, PRIME, Accelerated Approval), and orphan status, and cross-links to the drug it covers in DrugCore. ### GeneCore 43K+ /cores/genecore/records 43,000+ harmonized human gene records from HGNC, NCBI, UniProt, and Ensembl. Druggability and tractability assessments, target safety data, genetic constraint via gnomAD v4.0 (pLI, LOEUF), and cellular essentiality from CRISPR dependency screens. Each gene cross-links to the drugs that target it in DrugCore; follow those links to their trials, papers and approvals. ### PatentCore 170M+ /cores/patentcore/records Patent publications with English-preferred full text (title, abstract, claims, and description) plus bibliographic metadata (publication and application numbers, jurisdiction, kind code, family id), CPC/IPC classifications, inventors and assignees, four separate dates, and citation lineage. Search returns one publication per family, the most relevant member, and every record cross-links to the drugs it mentions in DrugCore and the papers it cites in BioMedCore. Tracks ## An API recipe for every track Inspiration, not coverage. Each command below was run against the live API and the responses are real (trimmed for width). Jump to your track: [1 · Longevity Biology × AI](#longevity-biology)[2 · AI Infrastructure — Nebius](#ai-infrastructure)[3 · Communication, Trust & Policy](#communication-trust)[4 · Healthspan & Clinical Translation](#healthspan)[5 · Open / Wildcard](#wildcard) Track ### 1 · Longevity Biology × AI Ageing clocks and biomarker inference, target triage for the biology of ageing, AI-supported drug repurposing, in-silico trials: defensible applications of AI to how we age. BioMedCoreGeneCoreDrugCore Ground an ageing-clock build in the actual methodology, ranked by journal quality and citations, cited to source. bash ``` curl "https://api.amass.tech/api/v1/cores/biomedcore/records\ ?query=epigenetic+clock+DNA+methylation+biological+age\ &minJournalQualityJufo=2&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · 4 records, JuFo ≥ 2 ``` TITLE JOURNAL YR CITED DNA Methylation Clocks in Aging: Categories, Causes… Molecular Cell 2018 670 Tracking the Epigenetic Clock Across the Human Life… J Gerontol A 2019 139 Epigenetic Aging: More Than Just a Clock When It… Cancer Research 2020 129 Retro-age: A unique epigenetic biomarker of aging… Aging Cell 2024 26 ``` Triage an ageing target in one call: SIRT1 comes back with clinical-precedent tractability, genetic constraint, and curated safety signals, each with the numbers to defend the pick. bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=SIRT1&limit=1" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · GeneCore SIRT1 ``` symbol SIRT1 location 10q21.3 LOEUF 0.527 pLI 0.90 targetClass Epigenetic regulator › Eraser › HDAC class III tractability small molecule: Advanced Clinical Structure with Ligand · High-Quality Ligand · Druggable Family safetyLiabilities • regulation of catalytic activity ToxCast ``` Also try For the repurposing half: DrugCore resolves “rapamycin mTOR inhibitor” to harmonized candidates with clinical stage (ridaforolimus at Approval, vistusertib and sapanisertib at Phase 2). Cross-link each hit into TrialCore for the trials that ran it and BioMedCore for the papers behind it. Track ### 2 · AI Infrastructure — Nebius A performant, scalable longevity-compute solution on the Nebius stack. Whatever you serve needs a grounded data layer behind it. Amass is one REST API with typed schemas and predictable limits, built to sit inside a pipeline. All six \*CoresOpenAPI Hydrate a feature store or vector index with harmonized records. Same query, same result, safe to cache hard. bash ``` curl "https://api.amass.tech/api/v1/cores/drugcore/records\ ?query=rapamycin+mTOR+inhibitor\ &drugType=SMALL_MOLECULE&limit=5" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · cache-friendly ``` NAME CHEMBL STAGE Ridaforolimus CHEMBL2103839 APPROVAL RG-7603 CHEMBL2331680 PHASE1 Vistusertib CHEMBL2336325 PHASE2 Palomid-529 CHEMBL2141712 PHASE1 AZD-8055 CHEMBL1801204 PHASE1/PHASE2 ``` Also try The plumbing that matters at scale: 60 requests / 60s with X-RateLimit-Remaining and X-Amass-Credit-Cost on every response, a flat { data } envelope, batch /lookup endpoints for external IDs, and a machine-readable OpenAPI spec at api.amass.tech/api/doc/openapi.json for code-gen and typed clients. Track ### 3 · Communication, Trust & Policy Products, tools, or campaigns that make longevity science understandable and credible: misinformation checkers, claim graders, responsible-adoption workflows. RegulatoryCoreBioMedCoreTrialCore What is metformin actually approved for? Sweep the unified FDA + EMA record: every authorization is type 2 diabetes, none is longevity. That gap is what a claim checker tests headlines against. bash ``` curl "https://api.amass.tech/api/v1/cores/regulatorycore/records\ ?query=metformin&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · FDA + EMA, unified ``` NAME AGENCY STATUS ACTIVE SUBSTANCE · INDICATION Riomet FDA WITHDRAWN_VOLUNTARY Metformin HCl · type 2 diabetes Jentadueto EMA ACTIVE Linagliptin;Metformin · type 2 diabetes Kazano FDA ACTIVE Alogliptin + metformin · type 2 diabetes Synjardy EMA ACTIVE Empagliflozin;Metformin · type 2 diabetes # no metformin authorization carries an ageing or longevity indication ``` Also try For the claim grader: BioMedCore carries isRetracted, journalQualityJufo (0–3), and citation counts on every record, enough to separate “one small study” from settled literature. TrialCore's hasResults and whyStopped distinguish tested from merely registered. Track ### 4 · Healthspan & Clinical Translation Clinician-facing tools, preventive-health workflows, and credible pathways from data or research to real-world healthspan applications. TrialCoreRegulatoryCore The human evidence for metformin as a geroprotector, in one call: two completed RCTs with posted results, and the next one still recruiting. bash ``` curl "https://api.amass.tech/api/v1/cores/trialcore/records\ ?query=metformin+aging+longevity&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · geroscience RCTs ``` TRIAL REGISTRY ID PHASE STATUS N RESULTS MILES NCT02432287 PHASE4 COMPLETED 16 posted ANTHEM NCT04264897 PHASE3 COMPLETED 166 posted AGE-M NCT06459310 PHASE2 RECRUITING 130 — (Padova) NCT01765946 PHASE4 COMPLETED 38 — ``` Healthspan is more than drugs. TrialCore covers the WHO ICTRP registries too, so exercise and nutrition interventions from Thailand to Mexico are one query away. bash ``` curl "https://api.amass.tech/api/v1/cores/trialcore/records\ ?query=sarcopenia+older+adults\ &overallStatus=COMPLETED&hasResults=true&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · beyond ClinicalTrials.gov ``` REGISTRY ID INTERVENTION N COUNTRY tctr TCTR20230703001 resistance exercise + whey protein 25 MY isrctn ISRCTN48485253 tai chi vs resistance exercise 120 MX isrctn ISRCTN17269170 social-media exercise programme 35 CN tctr TCTR20231005004 virtual-reality exercise 58 TH ``` Also try To keep a clinician tool inside the label: RegulatoryCore parses FDA labels and EMA SmPCs into addressable sections, so a workflow can quote the exact contraindication passage, id and all, instead of a summary. Track ### 5 · Open / Wildcard Propose your own challenge at the frontier of AI and longevity biotech. By definition you already know the problem; the whole engine is open, and the strongest entries usually chain \*Cores together. All six \*CoresCross-Core links Cross-Core links let one entity resolve across all six \*Cores, every hop cited to source. bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=SIRT1&include=referencesDrugCore&limit=1" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · one gene → the whole thread ``` SIRT1 → referencesDrugCore: the drugs that target it │ ├─ each drug → referencesTrialCore (the trials that ran it) ├─ each drug → referencesBiomedCore (the papers behind it) └─ each drug → referencesRegulatoryCore (FDA / EMA labels) gene → drug → trial → paper → approval, in one thread. all cited. ``` Playbook ## Nine ways to get further, faster 40% of the judging rubric is impact plus evidence and responsible use, and a demo that cites its sources scores on both. These patterns separate a build that lands from one that fights its own plumbing. 01 ### Build an agent, not a script. Let the model call the API in a loop: search broadly, filter, drill into a record, repeat. Our open-source starter agents (Python and TypeScript) ship exactly this loop; fork one and go. 02 ### Ground every claim in a record. Each result carries a stable Amass id and resolves back to its source: the PMID, NCT number, or patent record behind it. 03 ### Cross-Core is your edge. A gene resolves to the drugs that target it, the trials that ran them, the papers behind them, and the FDA/EMA labels that cover them. Follow the links, one call at a time. Treat an empty reference array as “none recorded,” not “no evidence.” 04 ### Query, do not scrape. PubMed, ClinicalTrials.gov, WHO ICTRP, ChEMBL, HGNC, UniProt, FDA & EMA are already normalised and cross-linked. Spend your weekend on the idea, not the ETL. 05 ### Narrow with filters. journalQualityJufo, phase, maxConstraintLoeuf, tractability, agency… tighter queries mean higher signal and fewer credits burned. Skip include=fulltext unless you need it. 06 ### Be nice to the rate limit. 60 requests / 60s, keyed by user + org. On a 429, read Retry-After and back off exponentially. Every response carries X-RateLimit-Remaining and X-Amass-Credit-Cost. 07 ### Read the data envelope. Every response is wrapped in { "data": … }; errors come back in a different shape under "error". Read results from data, and check for error before you parse. A batch lookup can even fail item-by-item. 08 ### Convert external IDs first. Get-by-id endpoints take Amass IDs only (AMBC\_, AMTC\_, AMDC\_, AMRC\_, AMGC\_). Holding PMIDs, DOIs, NCTs, ChEMBL, or gene IDs? Batch them through the /lookup endpoints, then fetch. 09 ### Cache, and mind pagination. Same query, same result. Memoise locally so a weekend of iterating does not re-spend credits. And there is no pagination: 300 results max per search, so narrow with filters instead of trying to page. Prototype in Claude ## Sketch your agent in Claude before you write app code. Connect the Amass MCP in Claude.ai, Claude Code, Cursor, ChatGPT, or Codex and query all six \*Cores in natural language. Explore the data, find the filters that work, then drop the same queries into the REST API in your build. In Claude Code, install the Amass SKILL.md and let the model wire up your data layer for you, a quick path from “what data is even in here?” to a working prototype. [Connect the MCP](https://amass.tech/mcp)Starter agent:[Python](https://github.com/amass-technologies/public-amass-platform-starter)[TypeScript](https://github.com/amass-technologies/public-amass-platform-starter-ts) Go further ## More ways to put the credits to work Not a coder on your team, or just want to move faster? Pick a lane. [ ### Fork an app from the gallery Nineteen starter apps (trial atlas, indication timeline, gene-to-approval) built on the same API. Fork one and remix it with your key. Browse the gallery](https://platform.amass.tech/app-gallery)[ ### No-code builds Paste your key into Lovable, Streamlit, or an n8n HTTP node and ship a filterable trial tracker or literature app. See examples](https://amass.tech/life-science-api)[ ### Claude Code + SKILL.md Install the Amass skill and let Claude Code build your Amass data layer, then iterate on the science instead of the SDK. Get the skill](https://amass.tech/mcp)[ ### Demo it live in Claude With the MCP connected, your agent’s answers come back cited in the chat, a clean, judge-friendly demo with zero deploy. How it works](https://amass.tech/mcp)[ ### Chain the \*Cores Strong entries fan out: gene → drug → trial → paper → label. Follow the reference fields and cite every hop. See the wildcard track](#tracks)[ ### Share on Discord Show what you are building with life-science AI builders, and tag the Amass team for a boost. Join now](https://discord.gg/sEGaBHMhWa) On the ground ### Meet the Amass mentors on-site [Henrik Jensen](https://www.linkedin.com/in/jensenhenrik/) (co-founder & CEO) and [Jörn Petersen](https://www.linkedin.com/in/petersenjoern/) (AI & data engineer) are your in-person mentors. Catch the Amass talk **Friday 16:45**, then find them in the **Saturday mentor rotations** for the promo code, API help, or to talk through your build. Off-site they answer on the event WhatsApp, Discord, LinkedIn, and email. [Meet the full Amass team →](https://amass.tech/about#team) [Henrik's LinkedIn](https://www.linkedin.com/in/jensenhenrik/)[Jörn's LinkedIn](https://www.linkedin.com/in/petersenjoern/)[Email](mailto:hello@amass.tech) ## Go build for longer, healthier lives Redeem your team’s $500 in credits, wire up a \*Core, and let an agent do the reading. Build for evidence, not theatre. [Claim your team’s $500 →](https://platform.amass.tech)[API reference](https://platform.amass.tech/api-reference)[Connect the MCP](https://amass.tech/mcp) One promo code per team, from the Friday talk or from Henrik and Jörn. Valid until 20 September 2026. Questions? [hello@amass.tech](mailto:hello@amass.tech) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Careers Careers # Build the research infrastructure for life science. We’re a small team in Copenhagen building a data engine across eight \*Cores, from 43M+ papers to 170M+ patents, that cites the records behind its answers. ## Why work here ### Work that gets used Life science researchers and investors use Amass to make decisions about drug development. ### Technical depth High-throughput data ingestion, semantic search at scale, cited answer generation and agent workflows, all running in production. ### Small team, big scope We're early. Everyone works across the stack and owns what they build. ### Copenhagen-based HQ in central Copenhagen. We value face-to-face time, focused work and shipping together. Open positions ## Roles open now [ Hybrid ### Unsolicited applications If no open role matches your profile today, apply anyway. Tell us what you'd work on at Amass and why. Read and apply→ ](https://amass.tech/careers/unsolicited-applications) ## Don’t see your role? If you’re a researcher, bioinformatician, designer or engineer who cares about this problem, send us an unsolicited application. [Send us your story →](https://amass.tech/careers/unsolicited-applications) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Data Engineer — Careers [←All roles](https://amass.tech/careers) Copenhagen # Data Engineer Build the data architecture and ML stack behind Amass, a **scientific intelligence platform** for life science R&D. Our customers are R&D teams in biotech, pharma and research institutions. With Amass they **triangulate evidence across datasets**, from patents and trials to internal reports and ELNs, and trace each decision back to its sources. The team is small. You'll work directly with the founders (CTO and CEO) on the data architecture, ML stack and applied intelligence layer at the center of **how Amass works**. Amass is backed by KOST Capital, Giant Ventures and EIFO. ## What we're building Scientific knowledge is scattered across patents, papers, regulatory filings and proprietary documents. Amass integrates them into a structured, cross-linked layer you can query. We **build our own domain-specific data pipelines and machine learning models** to capture subtle scientific signals. Users **triangulate across multiple sources**, trace the evidence behind any output and _reason_ with the documents. ## Your role We're hiring for ownership and initiative. You'll own core data systems and help decide how they're built. ### Your core tasks at Amass - Build fault-tolerant and reliable ingestion pipelines across diverse formats (APIs, PDFs, SharePoint, Excel, internal tools) - Design a scalable data foundation for low-latency scientific querying and reasoning - Maintain and optimize (vector) databases, embedding stores, and provenance tracking - Develop and maintain pipelines that serve (un)structured data, embeddings and metadata to downstream LLM workflows ### What you bring - _Have_ built a modern data foundation in a fast-moving team - Know how to ship data products that hold up in production while you iterate quickly - Like the mess of real-world data (scientific jargon, edge cases, legacy formats) - **Bonus:** You have experience working with data in life science, pharma, or another regulated industry - **Bonus:** You have experience building keyword, semantic or hybrid search systems Compensation is salary plus equity. ### How we work We are a flat organization and a small team. We value face-to-face time at our HQ in central Copenhagen. We work with high trust, solve problems together, try odd ideas and ship code without ego or status getting in the way. ### How to apply If we see a match, we'll reach out to schedule about two technical interviews. The last step is an onsite in our Copenhagen office where we'll work on a project together, discuss ideas, and meet the team. Apply ## Sound like you? Tell us about your background and what you'd bring to the data stack at Amass. [Apply for this role →](https://tally.so/r/m6zEjA) --- # Demand-gen / Growth marketer — Careers [←All roles](https://amass.tech/careers) Copenhagen, starting ASAP # Demand Gen / Growth Marketer Build the marketing function from scratch: campaigns, funnel analytics, creative automation and outreach to life science teams. At **Amass**, we build scientific intelligence infrastructure that helps life science teams find, connect and synthesize evidence across papers, trials, patents and regulatory sources. Backed by **Giant Ventures, KOST Capital and EIFO**, we're hiring our first **Demand Generation / Growth Marketer** to shape how we reach, convert and retain customers. You'll **build the marketing function from scratch**, from campaigns and funnel analytics to creative automation and outbound. ## What you'll do - Design and run full-funnel demand-gen experiments across channels (LinkedIn, email, communities, events, partnerships, etc.) - Build our early growth stack (CRM, attribution, automation and content ops), with AI tools you already use - Turn scientific stories into assets that perform (landing pages, newsletters, case studies, short videos) - Work with the founders and BD on positioning, messaging and outbound - Define the metrics you're measured on (activation, retention, LTV/CAC) and scale what works - Grow into a leadership role as the commercial team expands ## You might be a fit if you - Have 2–5 years' experience in growth, demand gen, or marketing at a SaaS, AI, or life-science-related company - Like data and experiments, and can tell a story scientists care about - Are fluent in modern tools (Attio, Figma, Webflow, Notion, GPTs, Claude etc.) - Build _with_ AI: you write prompts, automate and prototype instead of writing briefs - Want to grow into a Head of Marketing / GTM role and shape how Amass reaches its market ## What we offer - Work directly with the founders and our scientists and engineers - Ownership of a function you build from the start - Salary plus equity - HQ in central Copenhagen with a flexible, trust-based culture - High risk, high reward **When:**ASAP. We're collecting applications and will reply as soon as we can. Apply ## Sound like you? Tell us about yourself, your experience and how you'd build the marketing function at Amass. The form takes about 5 minutes. [Apply for this role →](https://3lbg9lg94m1.typeform.com/to/zRdWbONK) --- # Unsolicited applications — Careers [←All roles](https://amass.tech/careers) Hybrid # Unsolicited applications At **Amass**, we build scientific intelligence infrastructure for biotech and life science R&D. Our customers are R&D teams in biotech, pharma and research institutions. If no open role matches your profile today, apply anyway and tell us what you'd work on. ## What we value - Curiosity, creativity and a bias toward action - Technical, scientific or design skills, and the drive to keep learning - Comfort in a small, fast-moving team that runs on trust - Interest in hard problems whose results get used ## What you might work on - Designing and building AI/ML tools for scientific discovery - Designing interfaces that make complex data usable - Building scalable, fault-tolerant data systems - Shaping our product strategy and go-to-market efforts - Something else you can show us we need ## Why join us - Work directly with the founders - Build AI and data systems for biotech and pharma R&D - Salary plus equity - HQ in central Copenhagen with a flexible, trust-based culture ## How to apply Send us your CV, portfolio or anything else that shows your work, and tell us why you'd fit at Amass, even if your role doesn't exist yet. Apply ## Send us your story Tell us about yourself, what you've built and what you'd bring to Amass. [Send your application →](https://tally.so/r/wLrQay) --- # Contact Contact # Let’s talk. Send us enterprise questions, partnerships, security reviews or anything else, and we'll get back to you within one business day. Looking to see the product live? [Book a 20-min demo](https://amass.tech/demo). ## Email us Send questions about enterprise deployment, data, compliance or partnerships to [hello@amass.tech](mailto:hello@amass.tech) and the right person on our team will reply. For security or careers, use the topic-specific addresses on the right. [Email hello@amass.tech →](mailto:hello@amass.tech) ### Direct addresses General inquiries[hello@amass.tech](mailto:hello@amass.tech) Security questions[support@amass.tech](mailto:support@amass.tech) Careers[hello@amass.tech](mailto:hello@amass.tech) ### What to expect - ✓Response within 1 business day - ✓Routed to the right person on our team - ✓No marketing sequences Copenhagen, Denmark 🇩🇰 --- # Amass *Core — Eight *Cores, one retrieval engine Amass \*Core # Eight \*Cores, one retrieval engine. One query searches biomedical and scholarly literature, trials, drugs, genes, patents, FDA and EMA documents and the live web in parallel. Every claim in the answer carries the identifier of the record it came from, such as a PMID, an NCT ID or a patent number. [Get an API key](https://platform.amass.tech/api-keys)[Book a demo](https://amass.tech/demo) Architecture ## How a query fans out and comes back cited The engine sends one plain-language query to every relevant \*Core at once, merges the results and returns one JSON response, with a citation on each claim. 8 \*Cores App MCP API Amass \*Core architecture BioMedCore ScholarCore PatentCore RegulatoryCore DrugCore TrialCore GeneCore WebCore Provenance & citations External data providers - PubMed - PMC - Crossref - USPTO - EPO - WIPO - SureChEMBL - FDA - EMA - ClinicalTrials.gov - WHO ICTRP - HGNC - Ensembl - UniProt - ChEMBL - Web Connect your own data - SharePoint - OneDrive - Google Drive - Databricks - S3 / Blob - Network drives 500M+records in the registry 1. BioMedCore43M+ 2. ScholarCore233M+ 3. PatentCore170M+ 4. DrugCore22K+ 5. TrialCore1.2M+ 6. GeneCore43K+ Plus FDA + EMA regulatory records and the live web. [Details in the platform documentation →](https://platform.amass.tech/documentation/getting-started/overview) ### BioMedCore PubMed, PMC and conference abstracts BioMedCore ## 43M+ PubMed and PMC papers plus conference abstracts - Full PubMed and PubMed Central coverage with abstracts and full text - Conference abstracts from society meeting supplements - Citation counts, journal impact metrics, and quality signals - MeSH-enriched metadata for precise filtering BioMedCore Dapagliflozin in heart failure with preserved and mildly reduced ejection fraction: rationale and design of the DELIVER trial [PMID 34051124](https://pubmed.ncbi.nlm.nih.gov/34051124/)Solomon SD, McMurray JJV, et al.Eur J Heart Fail, 2021doi 10.1002/ejhf.2249 ### ScholarCore Scholarly papers across all disciplines ScholarCore ## 233M+ papers and preprints across every discipline - Journal articles, preprints and proceedings papers from outside biomedicine - Cross-disciplinary reach beyond the biomedical literature - Reaches work that has not yet appeared in BioMedCore ### PatentCore Life-science patents indexed PatentCore ## 170M+ patent filings from every major office: claims, families, and cross-links to drugs and papers - 16.8M publications across every major office: US, EP, WIPO/PCT, CN, JP, KR, and more - Full-text search over title, abstract, claims, and description - CPC/IPC classifications, assignees, inventors, and four dates (priority, filing, publication, grant) - Search collapses each family to one publication, so a landscape isn't flooded with near-duplicates - 12.5M patents linked to a drug in DrugCore; citation lineage and references into BioMedCore PatentCore GLP-1 analogues and derivatives [US-8815802-B2](https://patents.google.com/patent/US8815802B2)Novo Nordisk A/Sfiled 2010-12-16granted 2014-08-26 ### RegulatoryCore FDA & EMA documents indexed RegulatoryCore ## 7K+ FDA and EMA authorizations on one schema: approvals, labels, EPARs, and designations - FDA application records, drug labels, review documents, and approval letters - EMA EPARs, assessment reports, and SmPCs: European authorizations in full - Unified authorization status, plus Breakthrough, PRIME, and Accelerated Approval designations - Orphan tracking, cross-linked to the drug each authorization covers in DrugCore RegulatoryCore Kisunla (donanemab-azbt), Alzheimer’s disease [BLA 761248](https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=761248)FDAEli Lillyapproved 2024-07-02 ### DrugCore ChEMBL drugs & molecules indexed DrugCore ## 22K+ harmonized drugs and molecules: modality, clinical stage, and structure - ChEMBL-derived drug records with names, trade names, and synonyms, harmonized - Chemical structures with InChIKey and SMILES, plus modality classification - Highest clinical stage reached, from preclinical through approval - Parent/child relationships, cross-linked to trials, literature, and regulatory records DrugCore Semaglutide [CHEMBL2108724](https://www.ebi.ac.uk/chembl/explore/compound/CHEMBL2108724)ProteinApprovedOzempic · Wegovy · Rybelsus ### TrialCore Clinical trials indexed TrialCore ## 1.2M+ trials from ClinicalTrials.gov and international registries: status, endpoints, sponsor - ClinicalTrials.gov and international registries (590K+ and 610K+), with AI-enhanced search - Enrollment status, endpoints, and sponsor intelligence - Phase, indication, and comparator filtering - Trial-to-publication linking for evidence completeness TrialCore ESSENCE: semaglutide in non-alcoholic steatohepatitis [NCT04822181](https://clinicaltrials.gov/study/NCT04822181)Phase 3Active, not recruitingNovo Nordisk1,205 enrolled ### GeneCore Human genes indexed GeneCore ## 43K+ harmonized human genes, with druggability and constraint, cross-linked to DrugCore - HGNC, NCBI, UniProt, and Ensembl, harmonized on one schema - Druggability and tractability assessments, by modality and clinical precedent - Genetic constraint via gnomAD v4.0: pLI and LOEUF loss-of-function scores - Cellular essentiality from CRISPR dependency screens: genes required for cell survival - Each gene cross-links to the drugs that target it in DrugCore, and through those drugs to trials and papers GeneCore KRAS, KRas proto-oncogene, GTPase [HGNC:6407](https://www.genenames.org/data/gene-symbol-report/#!/hgnc_id/HGNC:6407)12p12.1UniProt P01116small molecule: approved drug ### WebCore Real-time web sources WebCore ## Real-time web sources, including company sites, press releases and conference news - Company sites, press releases and conference news - Current events that no fixed corpus has indexed yet - Cited the same way as every other source Capabilities ## What you get back from every query - ### Cited answers Citations sit inline, next to the claim they support. - ### Unified retrieval One query reaches all 8 \*Cores in parallel, and the results come back merged. - ### Agent execution Agents plan multi-step research, iterate on intermediate results and return structured output. - ### Enterprise controls Workspace roles, EU hosting, and single-tenant or bring-your-own-cloud deployment for regulated environments. - ### API and MCP access The REST API for your own code, the MCP for Claude, ChatGPT, Claude Code and Codex. Both serve six of the eight \*Cores. - ### Structured outputs Tables, JSON and tagged extracts your code can parse, as well as prose. ## Query the engine from your terminal The REST API serves six of the \*Cores. One request with your key returns typed JSON records with their source identifiers. [Get an API key](https://platform.amass.tech/api-keys)[Book a demo](https://amass.tech/demo) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Book a demo Book a call # 20 minutes with a founder about your setup You need to know whether Amass fits your environment before you buy. Pick a 20-minute slot and we'll go through fit, architecture, pricing and timeline: single-tenant deployment, SSO, procurement paperwork and onboarding. Exploring alone or with a small team? [Try free](https://amass.tech/try) No call needed. [Pick a time](https://calendar.google.com/calendar/u/0/appointments/schedules/AcZssZ2ewaDIvJzrmlou1BsIbqd2OBf-RHgcs-3YPYPR0wERWxAHGFy-pWtq64JYZbpQEpXPhvjf3y-2) What we'll cover - Site license, unlimited seats - SSO (SAML, OIDC, SCIM) - Role-based access - Single-tenant or BYOC deployment - Dedicated CSM and onboarding program How we handle your data - EU-hosted on Azure, with ISO 27001 and SOC 2 controls - Your data never trains a model - Every claim cited to its source record, with that record's ID ## What to expect ### 20 minutes, scoped A few minutes on your team, stack and constraints. The rest of the call is about how Amass fits. ### Architecture and data fit Single-tenant or BYOC, SSO setup and which \*Cores you need. If Amass doesn't fit your environment, we'll say so. ### Pricing, procurement, timeline Site-license sizing, procurement paperwork, onboarding cadence. You leave with the numbers and the steps to a signed agreement. --- # Book a technical discussion Talk to engineering # Talk to engineering about fit. You have questions from your own testing. Book 20 minutes with an Amass engineer on architecture, data access, your evaluation harness and the edge cases you found. No slides, no sales script. Looking for the commercial conversation instead? [Book a demo](https://amass.tech/demo) Procurement, pricing, deployment scope. [Pick a time](https://calendar.google.com/calendar/appointments/schedules/AcZssZ0V8yK5KGOpIweuB8sQrel6cLAyywjvKFFO-Stv1o607x9EoyNBl1XJj-xvXouF5I4qRC0Vl8Y4) What we'll cover - ✳\*Cores: BioMedCore and TrialCore - ✳MCP install + agent harnesses - ✳Raw API request and response shapes - ✳Evaluation on your own data - ✳Single-tenant or BYOC specifics How we handle your data - ✳EU-hosted on Azure, with ISO 27001 and SOC 2 controls - ✳Your data never trains a model - ✳Every claim cited to its source record, with that record's ID What to expect ### 20 minutes, engineering-led An Amass engineer on the call. Bring your repo, your evals and your odd edge cases; we work on whichever surface is closest to your decision. ### Architecture and data access How a request executes across BioMedCore, TrialCore and the retrieval engine. Auth, rate limits, citations, and what we will and won't return. ### Integration: MCP or API Which surface fits your harness. A live walkthrough of the MCP install or raw REST calls, whichever you plan to wire up. --- # Build on the Amass API at the DTU Skylab hackathon AI in Life Sciences, DTU Skylab, Wed 16 September 2026 # Turn life-science data into software in one evening. Amass hosts this hackathon with [Cursor Copenhagen and DTU Skylab](https://luma.com/cursor-jwp4). One REST API covers the six cross-linked \*Cores on the public API: 43M+ papers, 1.2M+ trials, 22K+ molecules, 43K+ genes, FDA and EMA authorizations, and patents. Every answer traces back to the record it came from. Copy the credit code, then pick one of the [four tracks](#tracks). You can also fork one of the [apps in the gallery](https://platform.amass.tech/app-gallery) and change the question it asks. The [build assistant](https://platform.amass.tech/assistant) writes the API calls with you. [Create your account →](https://platform.amass.tech)[Open the build assistant](https://platform.amass.tech/assistant) Doors open at 14:30 and the kickoff is at 15:00. Building runs from 15:30, demos start at 17:30, and prizes are announced at 18:00. The venue is the DTU Skylab Auditorium at Diplomvej 373D. Email [hello@amass.tech](mailto:hello@amass.tech) if you get stuck, or find us in the room. Your $50 credit code The code is worth $50 in Amass API credits. Create an account, redeem the code, then generate a key. It takes about two minutes. Then your first call bash ``` # every DTU paper on the topic, this year curl "https://api.amass.tech/api/v1/cores/biomedcore/records\ ?query=metabolic+engineering+yeast+cell+factory\ &institutionRors=04qtj9h94&minPublicationDate=2026-01-01" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK ``` Enzyme-constrained genome-scale model of Yarrowia lipolytica… Appl Microbiol Biotechnol Engineering and characterization of small-molecule transporters Trends Biotechnol Reusable and modular combinatorial libraries for S. cerevisiae Metab Eng pAo: a consensus genome-scale metabolic model for A. oryzae Comput Struct Biotechnol J Kinetic sampling shows the effect of medium composition… Microb Cell Fact # 04qtj9h94 is DTU's ROR id. Swap in your own ORCID with authorOrcids. ``` Start here ## Four ways in All four give you the same data with the same key. If you are not sure which one you want, start with the assistant. Start with this ### The Amass build assistant This is a chat grounded in the Amass documentation, built to help you write against the API. Describe what you are trying to do, such as finding every Danish Phase 2 trial in atopic dermatitis and the papers behind them. It answers with the endpoint, the filters and the field names, so you do not have to guess parameter names from a schema you have not read. [Open the assistant →](https://platform.amass.tech/assistant) [ ### The platform This is where your account, credits, and API keys live, alongside the full documentation, the interactive API reference, and a gallery of apps you can fork. Everything the assistant reads from is here. platform.amass.tech](https://platform.amass.tech)[ ### The Amass app This is the product itself. Ask a research question and get an answer with the records attached. Use it to scope an idea and see what the data can answer before you write a line of code. preview.amass.tech](https://preview.amass.tech)[ ### MCP Connect Amass to Claude, Cursor, or Codex and query all six \*Cores in natural language while you build. Use it to find the filters that work before you hard-code them. Connect the MCP](https://amass.tech/mcp) Apps to fork ## You do not have to start from an empty file Every app in the gallery runs on the same REST API, the same key, and the same cited records as the tracks below. All of them were built with [Lovable](https://amass.tech/life-science-api), so there is a route through this that involves no local setup at all. Fork one, point it at your key, and change the question it asks. [Browse the app gallery →](https://platform.amass.tech/app-gallery)[No-code recipes](https://amass.tech/life-science-api) platform.amass.tech/app-gallery Research #### From Gene to Approval One gene → every drug, trial, and FDA/EMA approval behind it. Research #### Graveyard & Garden 25 years of late-stage trials, halted or still alive. Research #### Clinical Trial Atlas Where a drug is tested, and which regions are enrolling. Research #### Indication Timeline A drug across its Phase-3 indications and approvals. Regulatory #### Transatlantic Divergence Desk FDA versus EMA outcomes, side by side. Clinical #### Pipeline Monitor A weekly digest of new Phase 2/3 trials by sponsor. Nineteen starter apps across research, regulatory, and clinical. Each one is a working answer to “what does a small app on this data look like?” The data ## Six \*Cores on the public API These are the six \*Cores that the REST API and the MCP serve today. They all take the same base URL and the same auth header, and they all return errors the same way. A paper, a trial, a molecule, a gene, an authorization and a patent can resolve to the same entity, so you can follow one thread instead of joining three exports by hand. ### BioMedCore 43M+ PubMed, PMC and conference abstracts /cores/biomedcore/records ### TrialCore 1.2M+ Clinical trials indexed /cores/trialcore/records ### DrugCore 22K+ ChEMBL drugs & molecules indexed /cores/drugcore/records ### RegulatoryCore 7K+ FDA & EMA documents indexed /cores/regulatorycore/records ### GeneCore 43K+ Human genes indexed /cores/genecore/records ### PatentCore 170M+ Life-science patents indexed /cores/patentcore/records The tracks ## Four tracks to choose from Pick one and make it yours. The first three come with a prompt you can paste into Cursor to get started, so you do not have to write the first line yourself. Every response shown below is a real record from the live API, trimmed to fit the column. [01 Literature review](#literature)[02 Prior art and FTO](#priorart)[03 Dataset analysis](#dataset)[04 Open track](#open) Track 01 ### Review the literature in your niche Pick a topic you care about and let an agent read everything published since you last looked. It keeps only the journals you would actually cite, and it writes a short digest with a PMID on every line. The version worth building is the one that runs again next Monday without you. BioMedCore localhost:3000/digest #### Monday digest · engineered probiotics 5 new since 1 May - Harnessing the microbiome for cancer therapyNat Rev Microbiol2026-06-01JuFo 3PMID 41486395 - Living drug carriers: microbial and bioengineered platforms redefining precision deliveryJ Control Release2026-08-24JuFo 3PMID 42636890 - The gut microbiome and allograft outcomesJ Clin Invest2026-08-17JuFo 3PMID 42606123 - Targeted and quantitatively modeled biologic delivery via dual-functional probiotic yeastJ Control Release2026-06-10JuFo 3PMID 41942055 - Developing spore-forming gut bacteria as model organismsTrends Microbiol2026-08-01JuFo 2PMID 42069481 One filter kept the journals worth citing. Every line carries the PMID it came from, so you can check any of them. Paste this into Cursor to start prompt ``` Build me a weekly literature digest. Topic: engineered probiotics for gut disease. Use the Amass API at api.amass.tech with my key in AMASS_API_KEY. Search BioMedCore for anything published in the last 30 days, keep only journals with journalQualityJufo 2 or higher, and write the result as markdown with the PMID on every line. Then save it as a script I can run again next Monday. ``` The call behind it bash ``` curl "https://api.amass.tech/api/v1/cores/biomedcore/records\ ?query=engineered+probiotic+living+therapeutic+gut+microbiome\ &minPublicationDate=2026-05-01&minJournalQualityJufo=2&limit=5" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · 5 records ``` TITLE JOURNAL DATE JUFO Harnessing the microbiome for cancer therapy Nat Rev Microbiol 2026-06-01 3 Living drug carriers: microbial and bioengineered platf… J Control Release 2026-08-24 3 The gut microbiome and allograft outcomes J Clin Invest 2026-08-17 3 Targeted and quantitatively modeled biologic delivery… J Control Release 2026-06-10 3 Developing spore-forming gut bacteria as model organisms Trends Microbiol 2026-08-01 2 ``` Also try You can point it at yourself instead of a topic. authorOrcids returns your own papers, and minCreateDate tells you what entered the index since your last run. That is the difference between a digest and a diff. Track 02 ### Check prior art before the idea becomes a company Skylab is full of projects that could be spun out, and the first honest question is what somebody has already claimed. Full-text search runs over the title, the abstract, the claims and the description. It also collapses each patent family to one publication, so a landscape comes back as five inventions rather than twenty filings of the same one. PatentCoreDrugCoreBioMedCore localhost:3000/prior-art #### Precision fermentation · recombinant milk protein 5 families · 4 assignees · US
Patent families by assignee, placed on a January 2019 to December 2021 priority-date axis, with each family’s priority date and forward-citation count.
Perfect Day Inc2019-04-22 · 2020-02-19 9 cited-by
Alpine Roads Inc2020-09-30 16 cited-by
Re Milk Ltd2021-05-12 1 cited-by
Fonterra Cooperative Group2021-06-24 0 cited-by
This is one row per family, not one row per filing. Twenty filings of the same invention collapse to the five things anyone claimed. Paste this into Cursor to start prompt ``` Help me check prior art before we take this idea further. Idea: making milk proteins by precision fermentation in yeast. Use the Amass API at api.amass.tech with my key in AMASS_API_KEY. Search PatentCore, keep US filings, and give me one row per patent family rather than one per filing. For each row show the assignee, the priority date and the citation count, sorted oldest first. Then tell me plainly which of them I have to read. ``` The call behind it bash ``` curl "https://api.amass.tech/api/v1/cores/patentcore/records\ ?query=precision+fermentation+recombinant+milk+protein\ &countryCode=US&limit=5" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · family-collapsed ``` PUBLICATION ASSIGNEE PRIORITY CITED-BY US-10988521-B1 ALPINE ROADS INC 2020-09-30 16 US-2023106635-A1 PERFECT DAY INC 2020-02-19 9 US-2024215599-A1 RE MILK LTD 2021-05-12 1 US-2024294589-A1 FONTERRA COOPERATIVE GROUP 2021-06-24 0 US-12458039-B2 PERFECT DAY INC 2019-04-22 0 # one row per family. priorityDate is the prior-art cutoff. ``` Also try Add include=claims to a hit and you see what is actually claimed rather than what the abstract implies. referencesBiomedCore walks back to the papers behind the filing. Track 03 ### Use AI to analyse a dataset Bring a dataset and let a coding agent do the first pass over it. Point Cursor at the file, ask what is in it, and make it show the code and the chart behind every claim it makes. Your own measurements work best here, and so does anything you can download in an hour. Remember that a chart is not a finding. Say what the data supports, and say where it is too thin. Bring your own dataCursor localhost:8888/analysis #### Your dataset, worked through in Cursor 4 steps you ▸Here is measurements.csv. Tell me what is in it, then tell me what is worth looking at. 1. Read the file and profiled it Columns, types, row count, missing values, and anything odd. 2. Plotted what actually varies One chart per column worth a second look, and nothing else. 3. Proposed three questions the data can answer Then said which of the three the data supports best. 4. Answered the strongest one With the code it ran and the chart behind the claim. The data is yours and the agent does the grunt work. Checking the result against published evidence is the optional last step. Paste this into Cursor to start prompt ``` Here is my dataset: measurements.csv. First tell me what is in it. Columns, types, row count, missing values, and anything that looks wrong. Then suggest three questions this data can actually answer, say which one is strongest, and answer that one. Show me the code you ran and the chart behind every claim, and tell me where the data is too thin to support a conclusion. ``` Also try When a result looks interesting, the next question is whether somebody has already found it. That is where the Amass API earns a place on this track rather than at the center of it. One BioMedCore search tells you whether your result is new or well known, and it hands you the PMIDs so a judge can check. Track 04 ### Open track: bring your own problem Bring the problem you already care about. This track has no shape we can describe in advance, and that is the point of it. One thing does hold across the entries that land, though: they chain several sources into a single answer, and they can show where each part of that answer came from. The example on the right is one shape that takes. Yours does not have to look anything like it. Any \*CoreCross-Core links localhost:3000/gene/GLP1R #### One question, answered across four \*Cores 4 calls · GLP1R LOEUF 0.494decile 1 pLI 0.986constrained PDB entries 703D solved Dependent lines 2 / 1,233not essential 1. GeneCoreGLP1RENSG00000112164 Family B GPCR · 6p21.2 · 463 aa · small molecule at Advanced Clinical 2. DrugCoreSemaglutideCHEMBL2108724 protein · 44 approved indications · Ozempic, Rybelsus, Wegovy 3. TrialCoreSELECTNCT03574597 Phase 3 · Novo Nordisk · N=17,604 · completed 2023, results posted 4. RegulatoryCoreWegovyAMRC\_3XhV… FDA 2021-06-04 and EMA 2022-01-06 · both authorizations active Gene, then drug, then trial, then approval. The links come with the records, so no hop in the chain is a guess. The call behind it bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=GLP1R&include=protein&limit=1" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · GeneCore GLP1R ``` symbol GLP1R ENSG00000112164 6p21.2 463 aa targetClass Membrane receptor › Family B GPCR › Glucagon-like peptide receptor tractability small molecule: Advanced Clinical · other clinical: Approved Drug constraint LOEUF 0.494 pLI 0.986 (LOEUF decile 1) essentiality not essential, 2 dependent lines of 1,233 screened structure 3D: yes · 70 PDB entries · Pfam PF00002, PF02793 ``` Also try include=referencesDrugCore returns the DrugCore ids for the molecules that target a gene. Each of those carries its own trial, literature and authorization links, so one call at a time gets you a long way. Follow the reference fields, and cite every hop. Playbook ## Five things worth knowing before 15:30 These are the things that separate a demo that lands from one that spends the evening fighting its own plumbing. 01 ### Two hours is the real constraint. Pick one endpoint and one output that somebody can see on a projector. A digest that actually runs beats a platform that almost does. 02 ### Let the build assistant write the calls. Describe what you want in plain language. The assistant answers from the documentation and hands back the request, the filters and the field names. That is most of the fiddly part done for you. 03 ### Ground every claim in a record. Each result carries a stable Amass id and a route back to its source. There are no invented PMIDs and no invented NCT numbers. That is the whole reason to query the data rather than ask a model from memory. 04 ### Narrow with filters, because there is no paging. A search returns at most 300 rows. Reach for minJournalQualityJufo, phase, maxConstraintLoeuf, assignee and date windows. A tighter query gives you better results and costs fewer credits. 05 ### Read the envelope and respect the limit. Results arrive under data and failures arrive under error, so check for an error before you parse. The limit is 60 requests per 60 seconds. If you get a 429, read Retry-After and wait before you try again. ## Go build something tonight Redeem the code, pick a track, and let the API do the reading. We would like to see what you made at 17:30. $50 in API credits Redeem it at platform.amass.tech once you have created your account. If it does not apply, find us in the room or email hello@amass.tech. [Create your account →](https://platform.amass.tech)[Build assistant](https://platform.amass.tech/assistant)[Try the app](https://preview.amass.tech)[Connect the MCP](https://amass.tech/mcp) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Events & webinars Events & webinars # Where to meet Amass Hackathons, webinars and talks where the team shows up, and a recording afterwards for every webinar. Most are free. Check this page for new dates, or follow Amass on Luma to hear about them first. [Follow on Luma →](https://luma.com/Amass) Upcoming ## Next up 22 Oct · Thu Next up WebinarAmass hosts ## How to do scientific work in Microsoft Copilot: cited research with Amass A live walkthrough of connecting Amass in Microsoft Copilot, then working through a trial landscape and the evidence around a target, with links to the ClinicalTrials.gov and PubMed records behind them. Thu 22 Oct 2026 · 14:00 CEST · Online [Register on Contrast →](https://watch.getcontrast.io/register/amass-technologies-how-to-do-scientific-work-in-microsoft-copilot)[Details →](https://amass.tech/resources/scientific-work-in-microsoft-copilot-kris-modig) Past ## Past events and recordings 1. 3–4 Oct HackathonAmass co-hostsEnded ### London AI x Science Hackathon A two‑day AI and science hackathon in London hosted by Iterate and futurebio.xyz. 3–4 Oct 2026 · 10:00 CEST · London [Event page ↗](https://luma.com/3iipivod) 2. 26–27 Sep HackathonAmass co-hostsEnded ### OFI x Amass — Vienna AI-for-Science Hack (HACK\_002) 24‑hour AI‑for‑Science hackathon in Vienna, 26–27 September 2026. 26–27 Sep 2026 · HOIV - Home of Innovation, Arsenalstraße 11, 1030 Wien, Austria [Event page ↗](https://luma.com/1ka4b1nm) 3. 16 Sep · Wed HackathonAmass co-hostsEnded ### Cursor Science Copenhagen: AI in Life Sciences Hackathon One evening at DTU Skylab turning life science data into working software, with students, PhDs and researchers from DTU and around Copenhagen. Hosted by Cursor Copenhagen with DTU Skylab and Amass. Bring a laptop. Wed 16 Sep 2026 · 15:00–19:00 CEST · DTU Skylab, Lundtofte [Event page ↗](https://luma.com/cursor-jwp4) 4. 11–13 Sep HackathonAmass speaksEnded ### Stockholm Longevity × AI Hack Stockholm Longevity × AI Hack, a public hackathon held Sep 11‑13 2026 at Epicenter Stockholm. 11–13 Sep 2026 · Epicenter Stockholm, Malmskillnadsgatan 44a [Event page ↗](https://luma.com/5b82vwsa) 5. 25 Aug · Tue WebinarAmass hostsEnded ### How to do scientific work in Claude: from general AI to domain expert A live walkthrough of connecting Claude to the Amass scientific data layer, then taking a real biologics asset evaluation end to end with every answer cited. Tue 25 Aug 2026 · 14:00 CEST · Online [Watch the recording →](https://amass.tech/resources/scientific-work-in-claude-kris-modig) 6. 30 Jul · Thu WebinarAmass hostsEnded ### Build with Lluís — BD & Competitive Intelligence Tools, From Idea to Internal App One real BD question, turned into a working internal tool live — using Claude connected to the Amass API, then Lovable for the app. Every number traceable to its source record. Thu 30 Jul 2026 · 14:00 CEST · Online [Watch the recording →](https://amass.tech/resources/build-with-lluis-bd-ci-tools) 7. 3–5 Jul HackathonEnded ### Building an AI Scientist Three-day hackathon in London where engineers, scientists and founders built new tools for drug discovery. Run by future.bio. 3–5 Jul 2026 · London [Event page ↗](https://luma.com/yw0c3upd) 8. 29 Jun · Mon WebinarAmass hostsEnded ### DrugCore + RegulatoryCore — Live Demo A live look at DrugCore and RegulatoryCore — linking a drug to its approvals, agency reviews and pivotal trials in one place. Mon 29 Jun 2026 · 14:00 CEST · Online [Watch the recording →](https://amass.tech/resources/drugcore-regulatorycore-live-demo) 9. 29 May · Fri WebinarAmass hostsEnded ### How to Use the Amass MCP in Your Scientific Work — Live Demo A live walkthrough of the Amass MCP inside Claude, ChatGPT and Cursor — how to connect it, three real research use cases, and Q&A. Fri 29 May 2026 · 14:00 CEST · Online [Watch the recording →](https://amass.tech/resources/amass-mcp-live-demo-webinar) Stay in the loop ## Hear about the next one first. Every Amass event goes on our Luma calendar before anywhere else. Follow it and Luma tells you when a new date lands. [Follow on Luma →](https://luma.com/Amass) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # How to do scientific work in Claude Scientific AI workflows # How to do scientific work in Claude Claude reasons well, but its conclusions are only as good as the evidence it can reach. Connect Amass and Claude searches six \*Cores, from 43M+ papers to 170M+ patents, and cites the records behind its answers. [Connect Amass to Claude](https://amass.tech/mcp)[See how it works](#how-it-works) --- What Claude gets ## Sourced evidence inside your Claude conversation. Claude's training data has a cutoff date and no access to live databases. With Amass connected, Claude queries six \*Cores during the conversation, and its answers link to the original sources. - Six \*Cores: 43M+ papers, 1.2M+ trials, 170M+ patents, 7K+ regulatory authorizations, 22K+ molecules and 43K+ genes - Inline citations that link to the original source - Structured filters: date, journal quality, retraction status, phase, sponsor - Multi-step search: Claude chains queries without you driving each step - Everything stays in the conversation, with no copying between tabs Get started ### Connect Amass to Claude Works with Claude Pro, Max, Team and Enterprise. Sign in once with your Amass account and you're set up. [See the install guide](https://amass.tech/mcp) Already connected? Add skills for specific workflows: [Browse the skill library →](#skills) --- What Amass covers ## Literature, trials and patents, with citations you can check. ### Biomedical literature Search 43M+ papers in BioMedCore, from PubMed and PubMed Central. Each result links to its source, so you can open the paper and check the passage. 43M+ papers ### Clinical trial data Search 1.2M+ trials in TrialCore, from ClinicalTrials.gov and the WHO ICTRP registries. Filter by condition, phase, sponsor, intervention and outcome to see what is being studied, where and by whom. 1.2M+ trials ### Patents & regulatory documents Search PatentCore and RegulatoryCore alongside the literature and trial data, to see what has been filed, approved or studied in an area. 170M+ patents --- --- How it works ## Connect once, then ask and read the sources. 01: Connect ### Add Amass to Claude once Four steps, ending in a one-time Amass sign-in. Requires Claude Pro, Max, Team or Enterprise. After that the connector queries Amass for you, with nothing else to configure. 02: Ask ### Type a question in plain language Ask about a condition, target, drug, trial, mechanism or company. Broad questions and specific ones about a sponsor or NCT number both work, with no special syntax. 03: Review ### Read the sources behind the summary Each result links to its record. Open the paper, inspect the citation, check the passage. Claude shows what it found, where it came from and where the gaps are. ## Start doing scientific work in Claude. Connect the Amass MCP once, and Claude’s answers come back with the sources they rest on. [Try free](https://amass.tech/try)[Connect Amass to Claude](https://amass.tech/mcp) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # How to use the Amass app — short video guides How to use Amass # Using the Amass app, in short video guides, one per feature. Real screen recordings of the app, under a minute each, with no narration. Pick a topic or jump to the video you need. [Open the app](https://app.amass.tech) Find the one you need 1. [01Ask your first question0:46](#ask-your-first-question) 2. [02Read the sources0:36](#read-the-sources) 3. [03See the sources as timelines0:44](#see-the-sources-as-timelines) 4. [04Steer the research0:27](#steer-the-research) 5. [05Ask about your own document0:37](#ask-about-your-own-document) 1. [06Build a report0:43](#build-a-report) 2. [07The plan and the map0:42](#the-plan-and-the-map) 3. [08Work with the tables in a report0:33](#work-with-the-tables-in-a-report) 4. [09Export to PDF0:22](#export-to-pdf) 5. [10Find and save your work0:31](#find-and-save-your-work) 1. [11Group your work in projects0:43](#group-your-work-in-projects) 2. [12Share a task with a link0:41](#share-a-task-with-a-link) 3. [13Set your theme and notifications0:24](#set-your-theme-and-notifications) 01 · Ask and read · 0:46 ## Ask your first question Type a research question in plain language. Amass shows every search it runs, writes a cited answer and suggests where to go next. 02 · Ask and read · 0:36 ## Read the sources What Amass cited and what it only read, filtered by data layer. Open any record and jump to the original source. 03 · Ask and read · 0:44 ## See the sources as timelines The same references as a trial landscape, an approval timeline and a literature timeline, each point a record you can open. 04 · Ask and read · 0:27 ## Steer the research Every answer ends with suggested next steps. Take one, or type your own follow-up, and the references grow with each step. 05 · Ask and read · 0:37 ## Ask about your own document Attach a PDF, Word, PowerPoint or text file and ask about it. Amass reads it, answers from it and cites it as your source. 06 · Build reports · 0:43 ## Build a report Describe the report you want. Amass scopes it, drafts a plan you can check, and builds a fully cited document you can read section by section. 07 · Build reports · 0:42 ## The plan and the map Before a report is built, edit its sections and sub-topics, and see on the map how the research converges on the final document. 08 · Build reports · 0:33 ## Work with the tables in a report Every table in a report is listed under Exhibits. Open one full size, click a cell's citation to read the record, and download it as CSV or Markdown. 09 · Build reports · 0:22 ## Export to PDF Every report exports to a PDF with a cover, the sections and every reference numbered at the end. 10 · Your workspace · 0:31 ## Find and save your work Pin the tasks you come back to, search and filter everything you have asked, and open any earlier report from one place. 11 · Your workspace · 0:43 ## Group your work in projects Create a project, file tasks into it and share the whole project with your team. 12 · Your workspace · 0:41 ## Share a task with a link Create a read-only link anyone can open without an account, see what your reader sees, and retire the link when you are done. 13 · Your workspace · 0:24 ## Set your theme and notifications Switch between light and dark, and let Amass notify you when a long-running answer or report is ready. ## Try it on your own question. Every answer comes back cited to the record it came from. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Amass for Microsoft Copilot | Scientific intelligence Integration # Amass scientific intelligence inside Microsoft Copilot Your scientists already use Copilot. Now Copilot can use Amass: through the Model Context Protocol (MCP), they research targets, review evidence and follow findings across publications, trials, patents and regulatory sources. [Get started](#get-started)[Read the announcement](https://amass.tech/resources/amass-microsoft-copilot) Use cases ## What you can research in Copilot 01 ### Targets and assets Investigate the evidence around a target or asset, from biological context to publications and clinical development. 02 ### Clinical and competitive landscapes Map the clinical programs in a research area, the assets in development and the evidence behind them. 03 ### Evidence across sources Work across publications, trials, patents and regulatory documents, and follow each finding back to its source. How it works ## How Amass works inside Copilot Amass structures and cross-links literature, clinical trials, patents, genes, drugs and regulatory documents. In Copilot it runs as a data source over the same MCP as in Claude and ChatGPT, so the answers keep their sources: trial answers link to ClinicalTrials.gov, paper answers to PubMed. ### Does your organization use Microsoft Copilot? Book a demo and we will show you how to turn on Amass in the Copilot your teams already use. [Book a demo](https://calendar.app.google/kBpdDJrS7h9bfqjHA) FAQ ## Questions about access What can I use Amass for? Target and asset research, evidence review, clinical and competitive intelligence, patent analysis and regulatory research. Does this replace the Amass app? No. Amass also works in its own app, through the API and through the MCP in Claude and ChatGPT. Microsoft Copilot is one more way in to the same scientific intelligence. What do I need to get started? Contact us to confirm the Amass account and Microsoft access your organization needs. Will I need help from our IT team? That depends on your organization's policies, which may require an administrator. We will help you identify the steps for your setup. Where can I read the launch announcement? [Read the Amass announcement.](https://amass.tech/resources/amass-microsoft-copilot) Get started ## Get started with Amass in Copilot Find out what your organization needs to enable Amass in Copilot: the access requirements and the setup steps. Tell us who you are and we will walk you through it. [Request access](https://calendar.app.google/kBpdDJrS7h9bfqjHA) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion Microsoft and Microsoft Copilot are trademarks of the Microsoft group of companies. --- # Amass MCP — Cited evidence inside your assistant Model Context Protocol # Cited biomedical evidence, inside Your assistant calls Amass search tools across six \*Cores and gets back papers, trials, drugs, genes, patents and regulatory records, each with its identifier. Works in Claude, ChatGPT, Codex, Claude Code and any MCP client. Get started with: [or any other MCP tool](#any-client) [or watch the CTO’s demo (1:37) →](#demo) APIBuilding with the API instead? [The Amass API: six \*Cores, cross-linked and cited.](https://amass.tech/life-science-api) Works with Claude Claude Code ChatGPT Codex Cursor Microsoft Copilot […and any other MCP-compatible tool](#any-client) Why MCP for evidence ## What the tool calls give your assistant. ### Filters built for evidence. The tools take filters for date, journal quality, retraction status and identifiers, so the assistant narrows the set before it reads. ### Searches that chain. The assistant runs a broad scan, filters it and opens specific records, without you driving each step. ### Citations back to sources. Every result carries the identifier and link of its record, so you can check a claim against its source. Install ## Connect Amass to your assistant. Pick your tool, add the connector and sign in once with your Amass account. Nothing else to configure: the assistant gets the access your plan gives you. Amass is listed in Claude's connector directory as **Amass Connector**. Find it in the app and connect in one click. Available on Claude Pro, Max, Team, and Enterprise. [Open Amass Connector in Claude](https://claude.ai/directory/amass) Press **Connect** on the listing and sign in once with your existing Amass account. That's the whole install. Your plan's access applies. See the step-by-step walkthrough (with screenshots) 1. Find Amass in Claude's connector directory. Use the button above — or in Claude, open **Settings → Connectors → Browse connectors** and search for **Amass**. 2. Open the listing, press Connect, and sign in with your Amass account. No URLs to paste, nothing to configure — the listing carries the setup. One-time sign-in, and your Amass plan's access applies. You'll be redirected to sign in with your Amass account. Your existing plan's access applies through the MCP. The skill ## Teach your assistant to use Amass well. Biomedical Evidence Scout is a small instruction pack: when to filter by journal quality, how to structure a landscape scan, when to open a specific record. It works in any assistant that reads the open SKILL.md format: Claude, Codex, GitHub Copilot, Gemini CLI. [Download .skill (Claude)](https://raw.githubusercontent.com/amass-technologies/public-skills/main/dist/amass-biomedical-evidence-scout.skill)[Download .zip (everything else)](https://raw.githubusercontent.com/amass-technologies/public-skills/main/dist/amass-biomedical-evidence-scout.zip) See [the Biomedical Evidence Scout in the Skills & Prompts library](https://amass.tech/skills/amass-biomedical-evidence-scout), with a KRAS G12C landscape briefing built from Amass data. How do I install the skill? Install every Amass skill in one step with `npx skills add amass-technologies/public-skills`. Or grab a file from the [amass-technologies/public-skills](https://github.com/amass-technologies/public-skills) repo: double-click the `.skill` on Mac to install it into Claude, or unzip the `.zip`into the tool's skills directory: - Claude Code: `~/.claude/skills/` - Codex: `~/.codex/skills/` - Gemini CLI: `~/.gemini/skills/` Try it ## Prompts to try once it's connected. The first tab is direct retrieval through the MCP. The second is multi-step work, where the skill guides the assistant. Clinical / translational ### Lp(a)HORIZON deep dive Can you pull up Lp(a)HORIZON and tell me the key trial details, whether results are available, and any papers linked to it? Returns: NCT ID, phase, sponsor, status, enrollment, countries, endpoints, posted results, and linked publications. Target / mechanism ### Recent TREM2 mechanisms What are the most important recent TREM2 mechanism papers, and what do they suggest about TREM2 as an Alzheimer's target? Returns: Ranked papers with journal, year, citations, PMID/DOI, model system, and core mechanistic claims. Competitive / BD ### RayzeBio active trials Which RayzeBio radiopharmaceutical trials are currently active, and what are the main assets, indications, and timelines? Returns: Active or recruiting trials, assets such as RYZ101/RYZ801/RYZ811/RYZ401, indications, phase, NCT IDs, enrollment, and completion dates. See it in action ## One query in Claude, with citations. Alex, Amass’s CTO, walks through a query in Claude. ClaudeChatGPTClaude CodeCodexAmass connected Which Lp(a)-lowering programs are in phase 3? amass`search_amass_trialcore_records`Completed `query: "lipoprotein(a) cardiovascular outcomes"``phase: "PHASE3"` [NCT04023552](https://clinicaltrials.gov/study/NCT04023552)Lp(a)HORIZON · COMPLETED · Novartis Pharmaceuticals [NCT05581303](https://clinicaltrials.gov/study/NCT05581303)OCEAN(a)-Outcomes · ACTIVE\_NOT\_RECRUITING · Amgen Two phase 3 programs: Lp(a)HORIZON (NCT04023552) has completed, and OCEAN(a)-Outcomes (NCT05581303) is active and no longer recruiting. Reply… The demo’s opening frame: the tool call, and the two trial records it returns. Microsoft Copilot ## Amass as a data source in Microsoft Copilot. Add Amass from the + menu in Copilot and ask your question there. The same MCP and the same sign-in run underneath, so the answers come from the same \*Cores and link to their sources. If Amass does not show up in your organization yet, we will help you set it up. [Amass for Microsoft Copilot](https://amass.tech/integrations/microsoft-copilot) Microsoft and Microsoft Copilot are trademarks of the Microsoft group of companies. Changelog ## Changes to the tools you call. Each release lists the tools, fields and filters it changed, newest first. 1. v10 ### GeneCore freshness filters and record dates September 2026 GeneCore search now takes the same freshness filters as BioMedCore, TrialCore and RegulatoryCore, and a fetched gene says when Amass last wrote it. - `get_amass_genecore_record` now also returns `lastUpdateDate` and `createDate` on each gene: when Amass last wrote the record, and when it first entered Amass. Search results do not carry them. - `search_amass_genecore_records` accepts `minLastUpdateDate` and `minCreateDate`, filtering on when Amass last wrote a gene or first ingested it, for "which genes changed since I last looked" and "what genes are new in Amass" questions. Any ingested change counts toward `minLastUpdateDate`: an HGNC edit, or a new target-intelligence, protein, RefSeq summary or drug-link value. Genes are refreshed weekly, and records without the date are excluded. 2. v9 ### Source links on records September 2026 Search and get results on every Core now include a `url` that links to the source of each record. - `search_amass_biomedcore_records` and `get_amass_biomedcore_record` return `url` next to `pmid` and `doi`, so an assistant can cite a publication with a working link instead of assembling one. It is null when a record has no PMID, DOI or PMCID. - The link target may change over time. `pmid` and `doi` are unchanged and remain the fields to build links to a specific external system from. - `search_amass_trialcore_records` and `get_amass_trialcore_record` return `url` next to `sourceUrl`. Today they carry the same registry link, except that a non-US trial with no registry link gets its WHO ICTRP record. `sourceUrl` keeps pointing to the registry even if `url` later moves. - `search_amass_drugcore_records` and `get_amass_drugcore_record` return `url` next to `chemblId`. - `search_amass_genecore_records` and `get_amass_genecore_record` return `url` next to `ensemblGeneId`. - `search_amass_regulatorycore_records` and `get_amass_regulatorycore_record` return `url` next to `sourceUrl`, with the same agency landing page today. `sourceUrl` keeps pointing to the agency even if `url` later moves. - `search_amass_patentcore_records` and `get_amass_patentcore_record` return `url` next to `publicationNumber`. 3. v8 ### Conference abstracts in BioMedCore September 2026 BioMedCore search and get now say that the Core holds conference abstracts as well as journal articles, and how to tell the two apart. - The BioMedCore tool descriptions and the server instructions name conference abstracts from society meetings as part of the Core, and state that such a record carries `Conference` in `publicationTypes` alongside NLM's own values. The data was already searchable; only the descriptions changed, so results and parameters are unaffected. - `search_amass_biomedcore_records` adds congress readouts that precede the full paper to what it is for, so an assistant reaches for BioMedCore on those questions instead of falling back to web search. - The headline count in both descriptions moves from 40M+ to 43M+, which is the corpus with conference abstracts included. Show all 10 releases 1. v10 ### GeneCore freshness filters and record dates September 2026 GeneCore search now takes the same freshness filters as BioMedCore, TrialCore and RegulatoryCore, and a fetched gene says when Amass last wrote it. - `get_amass_genecore_record` now also returns `lastUpdateDate` and `createDate` on each gene: when Amass last wrote the record, and when it first entered Amass. Search results do not carry them. - `search_amass_genecore_records` accepts `minLastUpdateDate` and `minCreateDate`, filtering on when Amass last wrote a gene or first ingested it, for "which genes changed since I last looked" and "what genes are new in Amass" questions. Any ingested change counts toward `minLastUpdateDate`: an HGNC edit, or a new target-intelligence, protein, RefSeq summary or drug-link value. Genes are refreshed weekly, and records without the date are excluded. 2. v9 ### Source links on records September 2026 Search and get results on every Core now include a `url` that links to the source of each record. - `search_amass_biomedcore_records` and `get_amass_biomedcore_record` return `url` next to `pmid` and `doi`, so an assistant can cite a publication with a working link instead of assembling one. It is null when a record has no PMID, DOI or PMCID. - The link target may change over time. `pmid` and `doi` are unchanged and remain the fields to build links to a specific external system from. - `search_amass_trialcore_records` and `get_amass_trialcore_record` return `url` next to `sourceUrl`. Today they carry the same registry link, except that a non-US trial with no registry link gets its WHO ICTRP record. `sourceUrl` keeps pointing to the registry even if `url` later moves. - `search_amass_drugcore_records` and `get_amass_drugcore_record` return `url` next to `chemblId`. - `search_amass_genecore_records` and `get_amass_genecore_record` return `url` next to `ensemblGeneId`. - `search_amass_regulatorycore_records` and `get_amass_regulatorycore_record` return `url` next to `sourceUrl`, with the same agency landing page today. `sourceUrl` keeps pointing to the agency even if `url` later moves. - `search_amass_patentcore_records` and `get_amass_patentcore_record` return `url` next to `publicationNumber`. 3. v8 ### Conference abstracts in BioMedCore September 2026 BioMedCore search and get now say that the Core holds conference abstracts as well as journal articles, and how to tell the two apart. - The BioMedCore tool descriptions and the server instructions name conference abstracts from society meetings as part of the Core, and state that such a record carries `Conference` in `publicationTypes` alongside NLM's own values. The data was already searchable; only the descriptions changed, so results and parameters are unaffected. - `search_amass_biomedcore_records` adds congress readouts that precede the full paper to what it is for, so an assistant reaches for BioMedCore on those questions instead of falling back to web search. - The headline count in both descriptions moves from 40M+ to 43M+, which is the corpus with conference abstracts included. 4. v7 ### MCP protocol revision 2026-07-28 September 2026 The server now speaks the 2026-07-28 MCP protocol revision alongside every earlier one, so assistants built on the newest MCP SDKs connect without an `initialize` handshake or a session. - Clients on the 2026-07-28 revision are served per request: no session id, and the client identifies itself on every call. - Clients on earlier revisions (2024-10-07 through 2025-11-25) keep working unchanged; the tools and their parameters are identical on both. 5. v6 ### Freshness filters, result limits, and data feedback September 2026 The search tools gained control over how many results come back and how recently Amass touched them. You can now optionally report data issues straight from the assistant to the Amass team. - Every search tool takes a `limit` (1–50) for how many results to return. The default stays at 10, so existing calls behave exactly as before. - BioMedCore, TrialCore, and RegulatoryCore search accept `minLastUpdateDate` and `minCreateDate`, filtering on when Amass last wrote a record or first ingested it — for "what changed since I last looked" and "what is new in Amass" questions, independent of publication, trial start, or authorization dates. - New send\_feedback MCP tool: report a wrong, missing, or outdated value in a record, a search that missed records that clearly exist, or a gap in the data model. It files the report with the Amass data team and is scoped to the six Cores — you name the Core(s) the issue concerns and, where it applies, the record id, the query you ran, and the ids you expected. It is not for web search or third-party results or context. Sending feedback is optional, and the tool returns only a receipt id rather than echoing the report back. 6. v5 ### International trials and PatentCore July 2026 TrialCore now spans non-US trials from the WHO ICTRP, and PatentCore adds patents as a new linked core. - TrialCore search and lookup now cover 1.2M+ trials — ClinicalTrials.gov plus non-US national registries via the WHO ICTRP (EUCTR, ChiCTR, JPRN, ISRCTN, and more). - Every trial record carries its source: a native registry ID (populated even for non-US trials, which have no NCT Id), the source registry, and a link to the trial on that registry. - The get\_amass\_trialcore\_record tool now accepts a registry ID (not just an Amass Id or NCT Id), so you can fetch non-US trials by their native registry id. - PatentCore search and lookup (preview) — patent publications with bibliographic metadata, full text (title, abstract, claims, description), CPC/IPC classifications, inventors/assignees, key dates, and citation lineage. Search the full text with filters for jurisdiction, classification, assignee/inventor, dates, and citation count. - PatentCore search collapses each patent family to one result, so an invention filed across many jurisdictions takes a single slot — the collapsed siblings' Amass IDs ride along on the kept record. - Patent records cross-link to DrugCore (the drugs disclosed in or linked to a patent) and BioMedCore (via references), and you can look a patent up directly by its publication number or Amass Id. 7. v4 ### GeneCore, target intelligence, and drug targets June 2026 GeneCore brings human genes into the linked cores with a full target-intelligence layer and DrugCore now carries the gene targets each drug acts on. - GeneCore search and lookup — harmonized gene records with approved symbols, names, synonyms, gene families, gene type (biotype), RefSeq functional summary, genomic location, external identifiers, and UniProt protein annotations. - Search filters for gene type/biotype, druggability, CRISPR essentiality, ChEMBL target class, clinical-precedent tractability (by modality and stage), curated safety liabilities, and gnomAD LOEUF constraint. - DrugCore records now return each drug's mechanisms of action — how it acts on each pharmacological target, its mode of action, and the target genes it hits. - Records link both ways: gene records point to the drugs (DrugCore) that target them, and drug records now carry their gene targets. 8. v3 ### RegulatoryCore and DrugCore June 2026 Coverage extends past papers and trials into the regulatory record and the drugs themselves — each core cross-linked to the rest. - RegulatoryCore search and lookup — FDA and EMA on one schema, from NDA/BLA approvals to EMA marketing authorizations. Filter by agency, molecule type, authorization status, authorization date, and orphan designation; cross-agency by default. - DrugCore search and lookup — harmonized drugs and molecules with trade names, synonyms, modality, clinical stage, and chemical structure. Filter by modality and clinical stage. - A new tool pulls the full text of a single regulatory document section — FDA labels and reviews, EMA SmPCs and EPARs. - Records now link across all four cores: a drug connects to its trials, its literature, and its regulatory authorizations. 9. v2 ### Better search, richer records May 2026 The two launch cores grew filters, optional full text, and the cross-links that tie papers and trials together. - BioMedCore search gained filters for publication date, journal quality (JuFo tier), retraction status, and author or institution — by name, ORCID, or ROR. - TrialCore search gained filters for phase, recruitment status, study type, intervention type, start date, and whether results have posted. - Fetching a publication can now return its full text on request, not just the abstract. - Records carry their cross-references — publication-to-trial links, plus what each record cites and what cites it. 10. v1 ### Launch — BioMedCore and TrialCore April 2026 The first release: two cores, each with a search tool and a direct-lookup tool. - BioMedCore — search and fetch across 39M+ PubMed and PMC publications, abstracts included. - TrialCore — search and fetch across 575K+ ClinicalTrials.gov trials, with protocols, sponsors, and endpoints. - Look any record up directly by identifier — PMID or DOI for papers, NCT Id for trials. 1. v7 ### MCP protocol revision 2026-07-28 September 2026 The server now speaks the 2026-07-28 MCP protocol revision alongside every earlier one, so assistants built on the newest MCP SDKs connect without an `initialize` handshake or a session. - Clients on the 2026-07-28 revision are served per request: no session id, and the client identifies itself on every call. - Clients on earlier revisions (2024-10-07 through 2025-11-25) keep working unchanged; the tools and their parameters are identical on both. 2. v6 ### Freshness filters, result limits, and data feedback September 2026 The search tools gained control over how many results come back and how recently Amass touched them. You can now optionally report data issues straight from the assistant to the Amass team. - Every search tool takes a `limit` (1–50) for how many results to return. The default stays at 10, so existing calls behave exactly as before. - BioMedCore, TrialCore, and RegulatoryCore search accept `minLastUpdateDate` and `minCreateDate`, filtering on when Amass last wrote a record or first ingested it — for "what changed since I last looked" and "what is new in Amass" questions, independent of publication, trial start, or authorization dates. - New send\_feedback MCP tool: report a wrong, missing, or outdated value in a record, a search that missed records that clearly exist, or a gap in the data model. It files the report with the Amass data team and is scoped to the six Cores — you name the Core(s) the issue concerns and, where it applies, the record id, the query you ran, and the ids you expected. It is not for web search or third-party results or context. Sending feedback is optional, and the tool returns only a receipt id rather than echoing the report back. 3. v5 ### International trials and PatentCore July 2026 TrialCore now spans non-US trials from the WHO ICTRP, and PatentCore adds patents as a new linked core. - TrialCore search and lookup now cover 1.2M+ trials — ClinicalTrials.gov plus non-US national registries via the WHO ICTRP (EUCTR, ChiCTR, JPRN, ISRCTN, and more). - Every trial record carries its source: a native registry ID (populated even for non-US trials, which have no NCT Id), the source registry, and a link to the trial on that registry. - The get\_amass\_trialcore\_record tool now accepts a registry ID (not just an Amass Id or NCT Id), so you can fetch non-US trials by their native registry id. - PatentCore search and lookup (preview) — patent publications with bibliographic metadata, full text (title, abstract, claims, description), CPC/IPC classifications, inventors/assignees, key dates, and citation lineage. Search the full text with filters for jurisdiction, classification, assignee/inventor, dates, and citation count. - PatentCore search collapses each patent family to one result, so an invention filed across many jurisdictions takes a single slot — the collapsed siblings' Amass IDs ride along on the kept record. - Patent records cross-link to DrugCore (the drugs disclosed in or linked to a patent) and BioMedCore (via references), and you can look a patent up directly by its publication number or Amass Id. 4. v4 ### GeneCore, target intelligence, and drug targets June 2026 GeneCore brings human genes into the linked cores with a full target-intelligence layer and DrugCore now carries the gene targets each drug acts on. - GeneCore search and lookup — harmonized gene records with approved symbols, names, synonyms, gene families, gene type (biotype), RefSeq functional summary, genomic location, external identifiers, and UniProt protein annotations. - Search filters for gene type/biotype, druggability, CRISPR essentiality, ChEMBL target class, clinical-precedent tractability (by modality and stage), curated safety liabilities, and gnomAD LOEUF constraint. - DrugCore records now return each drug's mechanisms of action — how it acts on each pharmacological target, its mode of action, and the target genes it hits. - Records link both ways: gene records point to the drugs (DrugCore) that target them, and drug records now carry their gene targets. 5. v3 ### RegulatoryCore and DrugCore June 2026 Coverage extends past papers and trials into the regulatory record and the drugs themselves — each core cross-linked to the rest. - RegulatoryCore search and lookup — FDA and EMA on one schema, from NDA/BLA approvals to EMA marketing authorizations. Filter by agency, molecule type, authorization status, authorization date, and orphan designation; cross-agency by default. - DrugCore search and lookup — harmonized drugs and molecules with trade names, synonyms, modality, clinical stage, and chemical structure. Filter by modality and clinical stage. - A new tool pulls the full text of a single regulatory document section — FDA labels and reviews, EMA SmPCs and EPARs. - Records now link across all four cores: a drug connects to its trials, its literature, and its regulatory authorizations. 6. v2 ### Better search, richer records May 2026 The two launch cores grew filters, optional full text, and the cross-links that tie papers and trials together. - BioMedCore search gained filters for publication date, journal quality (JuFo tier), retraction status, and author or institution — by name, ORCID, or ROR. - TrialCore search gained filters for phase, recruitment status, study type, intervention type, start date, and whether results have posted. - Fetching a publication can now return its full text on request, not just the abstract. - Records carry their cross-references — publication-to-trial links, plus what each record cites and what cites it. 7. v1 ### Launch — BioMedCore and TrialCore April 2026 The first release: two cores, each with a search tool and a direct-lookup tool. - BioMedCore — search and fetch across 39M+ PubMed and PMC publications, abstracts included. - TrialCore — search and fetch across 575K+ ClinicalTrials.gov trials, with protocols, sponsors, and endpoints. - Look any record up directly by identifier — PMID or DOI for papers, NCT Id for trials. FAQ ## Questions about the connector. Which assistants is the Amass MCP available in? Claude (Pro/Max/Team/Enterprise, listed in the connector directory), ChatGPT (Plus/Business/Enterprise/Edu, listed in the plugin directory), Claude Code and Codex CLI. In Microsoft Copilot, Amass is a data source you add from the + menu. Perplexity, Gemini CLI, Cursor, VS Code and other MCP clients connect with the generic connection details under Install, "Other clients". Is the connector approved by Anthropic, OpenAI, or others? Amass is listed in Claude's connector directory and in ChatGPT's plugin directory (the former app store), with one-click installs in both. Other MCP clients add it as a custom connector, which works the same way. Do I need an API key? No. You sign in once with your Amass account when you connect, and the MCP uses your plan's access. Does this work on ChatGPT Plus? Yes. The Amass MCP works on ChatGPT Plus, Business, Enterprise and Edu, with a one-click install from ChatGPT's plugin directory (the former app store). If your tenant blocks third-party apps, use the developer-mode fallback. What data does Amass have access to? Only the queries the assistant sends and the responses we return. The assistant itself runs on your provider's infrastructure. My company uses Microsoft 365. What are my options? Add Amass from the + menu in Microsoft Copilot under Change data sources, then sign in with your Amass account. If Amass does not show up there yet, or an administrator has to approve new sources, we will help with the setup. ## Cited biomedical evidence, inside Claude. Included on every plan, Free too, with one sign-in to your Amass account. Get started with: [or any other MCP tool](#any-client) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Install Amass MCP via ChatGPT developer mode — fallback Fallback install: ChatGPT developer mode # Install Amass MCP via ChatGPT developer mode. Use this if your tenant doesn't show Amass in ChatGPT's plugin directory yet. Some Enterprise tenants block third-party apps. The connector is the same; only the install path differs. [Back to the standard install](https://amass.tech/mcp#install-others) Available on ChatGPT Plus, Business, Enterprise, and Edu. This fallback path uses Settings → Apps → Developer mode on the ChatGPT web app — only needed when the Amass MCP app-store listing isn't visible to your tenant. 1. Open Settings → Apps and turn on Developer mode. From the ChatGPT sidebar, click your avatar → **Settings** → **Apps**. Developer mode is off by default — flip it on. ChatGPT will warn that it's an elevated-risk feature; that's expected. NoteMemory is automatically disabled in chats that use developer-mode connectors. Leave Enforce CSP in developer mode off unless your IT team requires it. Settings → AppsToggle Developer mode — **Create app** appears 2. Click Create app and fill in the connector details. A **New App** dialog opens. Use these values exactly. Description and icon are optional. Name **Amass MCP** MCP Server URL `https://mcp.amass.tech/mcp` Authentication OAuth Risk checkbox Tick “I understand and want to continue” `https://mcp.amass.tech/mcp` New App · BetaPress **Create** to continue 3. Sign in to Amass and allow access. ChatGPT redirects to `auth.amass.tech`. Sign in with your existing Amass account, then press **Allow access**. Your Amass plan's entitlements apply through the connector — no API keys to manage. NoteYou'll see the Amass mark, ChatGPT's avatar, and the email you're signed in as. Confirm it's the right account before allowing access. auth.amass.techPress **Allow access** 4. Confirm Amass MCP is connected. You're returned to ChatGPT and a green **“Amass MCP is now connected”** toast appears. Amass MCP also shows up under **Apps → Drafts**(private apps you've created in developer mode). ConfirmationToast appears in the top of the app 5. Start a new chat and attach Amass MCP. In the composer, click the + button and select **Amass MCP**from the menu. The composer gets a red Amass MCP chip and an orange ring — Amass is now in the conversation's tool stack. - Open a new chat (or reuse an existing one). - Click the `+` in the composer. - Pick **Amass MCP** from the dropdown. ComposerClick **+** → select **Amass MCP** You'll be redirected to sign in with your Amass account. Your existing plan's access applies through the MCP. --- # OpenEvidence EU Alternative EU evidence access # OpenEvidence left Europe. Amass is hosted here. Your evidence search still needs a home. Amass lets clinicians, researchers and life science teams search biomedical literature, clinical trials and regulatory sources, with a citation for each claim, on infrastructure hosted in the EU. Amass is for research and evidence review. It is not a diagnostic tool, a medical device or a replacement for clinical judgment. [Get your free tier — no card required](https://app.amass.tech)[See how it works](#how-it-works) --- Free access, no card required ## Free for clinicians and researchers in Europe The free tier is open to verified European clinicians, researchers, medical faculty and life science professionals. Sign up with your work or institutional email and start searching, with source links on every citation. - Search of biomedical literature, clinical trials, patents and regulatory documents - Inline citations linking directly to primary sources - EU-hosted infrastructure: your queries stay in Europe - No patient data required or intended - A monthly query allowance; upgrade only if you need more Free tier €0 no credit card required [Start Now](https://app.amass.tech) Sign up with your work or institutional email. Free monthly allowance included. Upgrade only if you need more. --- ### Important: read before using Amass is for research, evidence discovery and scientific intelligence. **It is not a medical device and does not provide diagnosis, treatment recommendations or clinical decision support.** Amass is not designed for point-of-care use and should not replace clinical judgment or direct patient care. Users are responsible for independently verifying all information and applying appropriate professional standards. Amass does not require or store patient data. Please do not enter personal health information or other patient-identifiable data. --- What Amass covers ## Literature, trials and patents, each cited to source ### Biomedical literature Search 43M+ papers from PubMed and PubMed Central. Each result links to the original source, so you can open the paper and check the passage. 43M+ papers ### Clinical trials Search 1.2M+ trials from ClinicalTrials.gov and the WHO ICTRP registries. Filter by indication, phase, sponsor, intervention and outcome to see what is being studied, where and by whom. 1.2M+ clinical trials ### Patents and regulatory documents Search patents and FDA and EMA regulatory documents alongside the literature and trial data, to see what has been filed, approved or studied in an area. 170M+ patents --- How it works ## Ask in plain language, get cited evidence 01 Ask ### Type a question in plain language Anything from a broad evidence question to a specific condition, target, drug, trial, mechanism or company. No query syntax or filters. 02 Retrieve ### Results come with citations Amass searches biomedical literature, trial records and patents, then returns a structured answer with links to the original sources. 03 Review ### Read the sources, not just the summary Amass shows what evidence it found, where it came from and where the gaps are, so you can check it and apply your own judgment. Amass returning cited evidence from the biomedical literature. Each statement links to its source. --- ## Get Started Today For European clinicians, hospital doctors, researchers and medical faculty. No credit card, no time limit. [Get Access](https://app.amass.tech) Sign up with your work or institutional email. --- # Pricing Pricing # Five plans and an API, on one engine. Free for trying Amass. Basic and Standard for one researcher, Business for teams of up to 10, Enterprise for sites, regulated teams and anything above 10 seats. Developers pay for the API by usage. Showing monthly prices. ## Free $0forever For students, curious researchers, and teams evaluating Amass. - 1 report + an included usage allowance - All eight \*Cores in the app - Every claim cited to source - MCP access: Claude, ChatGPT, Claude Code, Codex - 1 seat - Email + docs support [Try free](https://amass.tech/try) ## Basic $9per month For individual professionals, academics, and high-volume MCP users. - Everything in Free - 5 reports - A larger app + MCP allowance than Free - 1 seat [Get started](https://app.amass.tech/api/auth/sign-up?redirectTo=/settings/subscription?selectPlan=basic) Most popular ## Standard $39per month For biotech and pharma researchers, and VC analysts. - Everything in Basic - Unlimited reports, fair use - Our largest single-seat app + MCP allowance - Optional auto top-up when your allowance runs low - 1 seat [Get started](https://app.amass.tech/api/auth/sign-up?redirectTo=/settings/subscription?selectPlan=individual) ## Business $499per month For research teams of up to 10, sharing one workspace and allowance. - Everything in Standard - Up to 10 seats, shared workspace - Shared usage pool - Admin + billing console - Priority support + onboarding [Get started](https://app.amass.tech/api/auth/sign-up?redirectTo=/settings/subscription?selectPlan=large_team) [EnterpriseSites, regulated teams, above 10 seats: single-tenant, SSO, custom terms.See Enterprise](#enterprise) Building on top of Amass? [See API pricing](#api) or [compare all features](#compare) Enterprise ## For sites, regulated teams, and anything above 10 seats. Single-tenant deployment, SSO, procurement-grade paperwork and dedicated onboarding, scoped to your environment. [Book a demo](https://amass.tech/demo)[Talk to sales](https://amass.tech/contact) - Site license, unlimited seats - SSO (SAML, OIDC, SCIM) - Role-based access - Single-tenant or BYOC deployment - Dedicated CSM + onboarding program - Custom SLA, uptime guarantees, dedicated region API ## A usage-based REST API on the same engine An endpoint per \*Core, structured citations in every response, documented on platform.amass.tech. Same login as the app, separate billing stream. Pay as you go, billed monthly, no minimums. Volume discounts above 100k queries a month. [Get an API key](https://platform.amass.tech/api-keys)[Read the docs](https://platform.amass.tech/documentation/getting-started/overview) GET `https://api.amass.tech/api/v1/cores/trialcore/records?query=NCT04023552&limit=1` {"data": \[{"nctId": "[NCT04023552](https://clinicaltrials.gov/study/NCT04023552)","acronym": "Lp(a)HORIZON","phase": "PHASE3","overallStatus": "COMPLETED","sponsorName": "Novartis Pharmaceuticals",}\]} - Self-serve API key - MCP-compatible - EU-hosted Services ## Delivery help, from training to embedded engineers Three engagement levels, scoped to how much help your team needs: training, implementation, or engineers working inside your team. [Explore services →](https://amass.tech/services) - Enable: training and playbooks - Implement: scoped delivery - Forward Deploy: embedded team - Build-your-own-\*Core - API + architecture workshops - Scoped per engagement Every plan gets ## The same engine on every plan Every plan searches the same \*Cores with the same engine. Plans differ on reports, usage, seats and support, and never on data quality. All eight \*Cores All eight in the app, on every plan and every seat. The API and the MCP serve six of them. Cited to source Each claim carries its source: a DOI, NCT number, patent number or URL. The working trace Every \*Core it searched, the terms it used, and what it read but didn't cite. Fair use\* Unlimited reports on Standard and Business, under the fair-use note below. MCP access Connect Amass to Claude, ChatGPT, Claude Code, Codex or any MCP client, on every plan. EU-hosted Azure EU regions. Your queries never train a model. ISO 27001 + SOC 2 controls. \* Fair use covers unlimited reports. There is no hard cut-off: we contact heavy users to talk about their usage. Your usage allowance is a separate limit and pauses when it runs out (see the FAQ). Compare plans ## Every feature, side by side. The full matrix, API column included; the plan cards above cover most decisions. Scroll for more plans | Feature | Free | Basic | Standard Most popular | Business | Enterprise | API | | --- | --- | --- | --- | --- | --- | --- | | Data + answers | | \*Cores | All 8 | All 8 | All 8 | All 8 | All 8 | 6 | | Inline citations to source | | | | | | | | Structured JSON responses | | | | | | | | Each claim traced to its source record's ID | | | | | | | | No model training on your queries | | | | | | | | Reports + usage | | Reports | 1 | 5 | Unlimited, fair use | Unlimited, pooled | Custom | — | | Included usage allowance | Included | Larger than Free | Largest single-seat | Pooled across seats | Custom | Usage-based | | Ask and Build in the app | | | | | | — | | Working trace on every answer | | | | | | — | | MCP access | | | | | | | | Programmatic API | — | — | — | — | Add-on | | | Seats + collaboration | | Seats | 1 | 1 | 1 | Up to 10 | Site license | — | | Shared workspace | — | — | — | | | — | | Shared usage pool | — | — | — | | | — | | Admin + billing console | — | — | — | | | | | Security + compliance | | EU hosting | | | | | Dedicated region | | | SSO (SAML / OIDC / SCIM) | — | — | — | — | | Add-on | | Role-based access | — | — | — | — | | Add-on | | Single-tenant or BYOC deployment | — | — | — | — | | — | | Custom SLA + uptime guarantees | — | — | — | — | | Volume plans | | Support | | Support channel | Docs + email | Docs + email | Docs + email | Priority + onboarding | Dedicated CSM | Docs + email | | Onboarding | — | — | — | Group onboarding | White-glove program | — | | Response SLA | — | — | — | 48h | Custom | 48h | | Billing | | Monthly or annual billing | Free forever | | | | Annual contract | Monthly, usage-based | | No credit card to start | | — | — | — | — | | FAQ ## Questions about plans and billing Do I need a credit card to sign up for Free? No. Free needs an account but no card, and it never converts to a paid plan automatically. Is MCP available on the Free tier? Yes. MCP access is on every plan, Free included. Connect Amass to Claude, ChatGPT, Claude Code, Codex or any MCP client. What counts as a report? A report is a document Amass writes from a plan: in Build you review the sections, press Run plan, and get a report with numbered citations and its working trace. Answers in Ask and MCP calls are not reports. They draw on your usage allowance, and the app shows your balance. What happens when I run out of my usage allowance? Ask, Build and MCP calls pause until the allowance resets on your billing date, and reports pause with them, since each starts with a planning step that uses the allowance. On Standard, auto top-up under Settings › Subscription charges your saved card before you run out, at the amount, trigger and monthly cap you set. The monthly allowance is spent first; topped-up usage never expires. Fair use is a separate limit: it covers unlimited reports, has no cut-off, and means we contact heavy users. What's the difference between Basic, Standard, and Business? Basic and Standard are single-seat plans. Basic includes 5 reports and a larger usage allowance than Free. Standard has unlimited reports under fair use and our largest single-seat allowance, for people who use Amass daily. Business is up to 10 seats in a shared workspace, with one pooled allowance and priority support. Above 10 seats, it's Enterprise. Can I switch plans later? Yes. Upgrades apply immediately, downgrades at the next billing cycle. Basic, Standard and Business have no contract. How does billing work for the Amass API? The API is billed separately, by usage, every month. Same login, separate billing stream, so a Business subscription and an API account run side by side without double-charging. Can you build custom agents or integrations for us? Yes. Amass Professional Services gives you dedicated developer time to build agents, integrations and workflows for your team. Contact us to scope one. Do you offer academic or non-profit pricing? Yes. Email hello@amass.tech and we'll work something out. ## Start with Free. Move when you’re ready. No card required, and Free never turns into a paid plan on its own. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Product Product # One research engine. Three ways to work. Amass is one engine over eight \*Cores, from 43M+ papers to 170M+ patents. The app searches all eight and cites every claim; the API and the MCP serve six of them to your code and your assistant. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) [01App Ask a research question and watch it work: every claim cited, and what it read but didn’t cite shown too.](#cr-panel-app) [02MCP Amass inside the tools you already use: Claude, ChatGPT, Claude Code and Codex, over MCP.](#cr-panel-mcp) [03API Build on the engine: typed, cited responses you can program against.](#cr-panel-api) [See the App →](https://amass.tech/app) The same question three ways: the app with its trace and citations, an assistant calling the Amass tool over MCP, and the API returning the record. How Amass works ## Follow the evidence from record to answer Amass searches life science records, follows the links between them and returns answers with sources you can check. 1. [**01**Data: life science recordsLife science sources held as records.](#product-story-board-0) 2. [**02**Connected: follow a record’s linksLinks join trials, drugs and papers.](#product-story-board-1) 3. [**03**Sourced: check each claimEach claim points to its source.](#product-story-board-2) Every answer from the app, the MCP or the API carries the identifiers of its sources. Check a claim against its DOI, NCT ID, ChEMBL ID, patent number or regulatory application number. TrialCore1 Lp(a)HORIZON [NCT04023552](https://clinicaltrials.gov/study/NCT04023552) DrugCore2 pelacarsen [CHEMBL4298202](https://www.ebi.ac.uk/chembl/explore/compound/CHEMBL4298202) BioMedCore3 Early health technology assessment [10.1016/j.jacl.2024.08.012](https://doi.org/10.1016/j.jacl.2024.08.012) Answer Lp(a)HORIZON [1](https://clinicaltrials.gov/study/NCT04023552) tests pelacarsen [2](https://www.ebi.ac.uk/chembl/explore/compound/CHEMBL4298202) for lowering lipoprotein(a), the subject of an early health technology assessment [3](https://doi.org/10.1016/j.jacl.2024.08.012). 43M+ BioMedCore 233M+ ScholarCore 170M+ PatentCore 7K+ RegulatoryCore 22K+ DrugCore 1.2M+ TrialCore 43K+ GeneCore Live WebCore [Deep dive into Amass \*Core](https://amass.tech/core) Access ## Build it, connect it, or use it Three ways into the same engine. Pick one, or combine them. [ ### The Amass app No code. Ask a question, get a cited answer you can check. Try free ](https://app.amass.tech/api/auth/sign-up)[ ### The Amass API Typed JSON from six \*Cores, every record cited. Build research into your product. GET /api/v1/cores/trialcore/records?query=NCT04023552&limit=1 { "data": \[{ "nctId": "NCT04023552", "acronym": "Lp(a)HORIZON", "phase": "PHASE3" }\] } Get an API key ](https://platform.amass.tech/api-keys)[ ### Amass MCP A remote server and one sign-in. Claude, ChatGPT, Claude Code and Codex query Amass with citations. { "mcpServers": { "amass": { "type": "http", "url": "https://mcp.amass.tech/mcp" } } } Set up in two minutes ](https://amass.tech/mcp) Use cases ## What you can build with Amass From a single question to a full evidence review, on the same engine. ### Due diligence Build a competitive landscape brief from patents, trials and literature. ### Regulatory intelligence Ask what the FDA and EMA records say and get answers built from the labels and reviews. ### Literature review Cross-reference 43M+ biomedical papers with patent and trial data in one query. ### Internal tools Embed cited answers in ELN sidebars, dashboards and compliance workflows through the API. [See all use cases](https://amass.tech/use-cases) ## Pick your path Start free in the app or with an API key. Book a demo when your team needs a deployment scoped to it. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Resources Resources # Guides, insights, and documentation Guides and API references for getting more out of Amass, and a newsletter on life science research intelligence. [API overview →](https://platform.amass.tech/documentation/getting-started/overview)[BioMedCore reference →](https://platform.amass.tech/documentation/cores/biomedcore)[TrialCore reference →](https://platform.amass.tech/documentation/cores/trialcore)[Quickstart →](https://platform.amass.tech/documentation/getting-started/quickstart)[Events & webinars →](https://amass.tech/events) 17 articles[Webinars and recordings live under Events →](https://amass.tech/events) [ NewsFeatured ## Amass and SciSure partner to bring scientific AI into laboratory workflows worldwide The partnership connects Amass scientific intelligence with SciSure serving more than 550,000 scientists, EHS and LabOps users across 55,000 laboratories worldwide. Read article →Oct 2026 · 3 min read ](https://amass.tech/resources/amass-scisure-integration) [ Amass · Resource Every FDA and EMA Authorization for One Substance Use Case ### Every FDA and EMA Authorization for One Substance Semaglutide holds ten FDA and EMA authorizations under five invented names. RegulatoryCore returns all ten from one search on the substance. Oct 202612 min read ](https://amass.tech/resources/every-fda-and-ema-authorization-for-one-substance)[ News ### Amass brings scientific intelligence infrastructure to Microsoft Copilot New integration connects Amass’ scientific intelligence infrastructure with Microsoft Copilot, bringing interconnected life science knowledge directly into the everyday workflows of scientists and life science organizations. Sep 20265 min read ](https://amass.tech/resources/amass-microsoft-copilot)[ Amass · Resource CDH17 at AACR and ASCO 2026: 26 molecules and one journal paper with patients Use Case ### CDH17 at AACR and ASCO 2026: 26 molecules and one journal paper with patients BiomedCore now includes conference abstracts. For CDH17, the AACR and ASCO 2026 abstracts list 26 molecules, and the journals hold one paper with patients. Sep 202610 min read ](https://amass.tech/resources/cdh17-aacr-asco-2026-conference-abstracts)[ Product ### The new Amass app is live The largest update to Amass since the app first launched. Ask across all eight \*Cores, steer the research as it runs and trace every claim to its source. Sep 20267 min read ](https://amass.tech/resources/the-new-amass-app-is-live)[ News ### Amass × BASE life science: taking scientific AI to enterprise scale Amass and BASE life science are partnering to take scientific AI from individual use cases to enterprise-scale implementation in life sciences. Aug 20262 min read ](https://amass.tech/resources/amass-base-life-science-partnership)[ Customer ### Killing drug targets early: how Apoptosis AI won the London hackathon Two winning teams from the London "Building an AI Scientist" hackathon built on the Amass Cores, and both were still building on them more than a month later. Here is what they built and why they reached for a curated, linked data instead of a search tool. Aug 202611 min read ](https://amass.tech/resources/apoptosis-ai-target-assessment)[ ### What the Building an AI Scientist London Hackathon taught us about Tech x Bio What we learned from 50 builders and what it taught us about building successfully in Tech x Bio. Jul 20265 min read ](https://amass.tech/resources/building-ai-scientist-tech-x-bio)[ Amass · Resource From molecule to market: introducing DrugCore and RegulatoryCore ### From molecule to market: introducing DrugCore and RegulatoryCore Introducing DrugCore and RegulatoryCore: drugs and FDA/EMA approvals, harmonized and linked to the trials and literature you already query. Jun 20267 min read ](https://amass.tech/resources/introducing-drugcore-regulatorycore)[ Technical ### Introducing GeneCore: One Link From a Drug to Its Target Biology Introducing GeneCore, one canonical record per gene, plus mechanism of action on every DrugCore record. See how the two connect a drug to its target biology in a single API link. May 20267 min read ](https://amass.tech/resources/introducing-genecore-mechanism-of-action)[ Amass · Resource Eval First, Model Second: Inside the TrialCore Iteration Loop Technical ### Eval First, Model Second: Inside the TrialCore Iteration Loop A look at the evaluation infrastructure behind one of Amass's core retrieval services. How we build the datasets, why we trust the labels, and the loop we run to actually move metrics. May 20269 min read ](https://amass.tech/resources/eval-first-model-second-trialcore-iteration-loop)[ Technical ### Comparing 4 biomedical MCPs using a fixed agent harness Aggregate results and a worked NSCLC case study from a four-server biomedical MCP benchmark (Amass, PubMed, Consensus, Paperclip). May 202613 min read ](https://amass.tech/resources/comparing-4-biomedical-mcps-claude-code)[ Product ### Why biomedical AI fails at retrieval before it fails at reasoning A 10-question benchmark across Amass BioMedCore, Claude with PubMed, and ChatGPT with web search. Retrieval, not reasoning, is biomedical AI's bottleneck. Apr 202612 min read ](https://amass.tech/resources/why-biomedical-ai-fails-at-retrieval)[ Product ### Monitor description and examples Mar 20261 min read ](https://amass.tech/resources/monitor-description-and-examples)[ Product ### Introducing Planner: A New Way to Guide Your Research Mar 20262 min read ](https://amass.tech/resources/introducing-planner-a-new-way-to-guide-your-research)[ Product ### Introducing \*Core: The Intelligence Architecture Behind Amass All six external data sources are being renamed to a unified \*Core naming convention — BioMedCore, ScholarCore, PatentCore, RegulatoryCore, TrialCore, and… Mar 20263 min read ](https://amass.tech/resources/introducing-core-the-intelligence-architecture-behind-amass)[ Product ### Global Data Context and Placeholder Positioning Enhanced accessibility and configurability of report placeholders and global data context in the Amass platform improve user experience. Feb 20262 min read ](https://amass.tech/resources/global-data-context-and-placeholder-positioning) ## Subscribe to the Amass newsletter Life science intelligence, product updates, and research trends, in your inbox twice a month. No spam. Unsubscribe anytime. We don't share your email. ## Building on the API? Full API documentation and reference at platform.amass.tech [Go to docs →](https://platform.amass.tech/documentation/getting-started/overview)[API overview](https://amass.tech/life-science-api) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Services — Training, implementation and embedded engineering Amass Services # From first use case to production deployment Pick how you use Amass, then how much help you want. Use the app, integrate the API, or have our engineers design, build and run a scientific intelligence workflow in your environment. [Book a demo](https://amass.tech/demo)[Try free](https://amass.tech/try) Engagement scopeIllustrative ObjectiveGoverned target-research workflow, in productionAccessAmass API, internal data connectedEngagementImplement, with Team Enablement includedPhasesAlign → Design → Prove → DeployDeliverablesRunning integration, evaluation report, handover 43M+biomedical papers indexed 170M+patents, searchable in one schema 1.2M+clinical trials, updated daily 6\*Cores behind one API, eight in the app How do we use Amass? ## Choose how Amass fits into your organization Start with where Amass should live. The level of support you need is the second question. ### Amass App A research workspace where your team asks questions and gets answers with every claim cited to its source. Best for: Teams that want to start now, without building an integration. [Explore the App →](https://amass.tech/app) ### Amass API REST access to Amass search, retrieval and records across six \*Cores. Best for: Product teams and internal engineering teams building their own workflows, agents or applications. [Explore the API →](https://amass.tech/life-science-api) ### Embedded Amass Amass integrated into an internal platform, customer-facing product or operational workflow. Best for: Organizations that want Amass inside an existing product or infrastructure layer. [Discuss an embedded deployment →](https://amass.tech/demo) ### The two questions, combined Access is one axis, support is the other. Tap any pairing to see what the engagement looks like. App API Embedded Amass Enable Implement Forward Deploy Amass API × Implemente.g. API implementation and migration We build the integration (data mapping, auth, testing) and hand over documented, running code. How does Amass help us succeed? ## Choose the level of support you need Three levels, from training your team to engineers embedded with it. We combine and scope them per engagement. ### Enable Get your team using Amass on its own work. - Stakeholder and use-case discovery - Team onboarding - Role-specific training - Prompt and workflow development - API and architecture workshops - Governance and usage recommendations - Office hours and adoption support What's included (7) - Stakeholder and use-case discovery - Team onboarding - Role-specific training - Prompt and workflow development - API and architecture workshops - Governance and usage recommendations - Office hours and adoption support Typical outcome: A trained team, prioritized use cases and an adoption plan. ### Implement Move a defined use case into a working production workflow. - Technical discovery and solution design - Data-source and workflow mapping - API integration - Authentication and access configuration - Internal data integration - Application or agent implementation - Testing, evaluation and acceptance criteria - Deployment support - Technical documentation and handover What's included (9) - Technical discovery and solution design - Data-source and workflow mapping - API integration - Authentication and access configuration - Internal data integration - Application or agent implementation - Testing, evaluation and acceptance criteria - Deployment support - Technical documentation and handover Typical outcome: A deployed workflow, integration or application that your team can own and operate. ### Forward Deploy Embedded Amass engineers join your team and deliver with it. - Embedded forward-deployed engineer or team - Product and workflow design - Custom agents, applications and interfaces - Scientific evaluation frameworks - Data architecture and ontology work - Production deployment - Monitoring and optimization - Ongoing roadmap and operational ownership - Training and internal capability transfer What's included (9) - Embedded forward-deployed engineer or team - Product and workflow design - Custom agents, applications and interfaces - Scientific evaluation frameworks - Data architecture and ontology work - Production deployment - Monitoring and optimization - Ongoing roadmap and operational ownership - Training and internal capability transfer Typical outcome: A production system delivered and evolved jointly with Amass. The combination and scope go into an order form or statement of work. Larger subscriptions may include enablement; implementation and forward-deployed work are usually scoped separately. The strategic engagement ## Build your own \*Core Turn your organization's knowledge into a dedicated intelligence layer. A custom \*Core combines selected external data, internal knowledge, scientific workflows and domain logic. We design, build and deploy it with your scientific, product and technical teams. What goes in Amass datasets and intelligence services Licensed or proprietary external data Internal documents, reports and structured data \*Corestructured by your taxonomies, workflows and evaluation criteria What comes out Custom agents, applications and monitoring Governed research and decision workflows Access controls, governance and full traceability Possible \*Core types ### Company Core An intelligence layer across internal research, assets, programs and external evidence. ### Therapeutic Area Core Structured knowledge for a disease area, modality or biological domain. ### Asset Core A continuously updated view of a target, molecule, program or portfolio. ### Regulatory Core Internal and external regulatory knowledge organized around products and markets. ### Customer Core A scientific intelligence capability embedded into a platform or customer-facing product. ### Content Core Controlled access to proprietary content through search, workflows and applications. Possible \*Core types (6) ### Company Core An intelligence layer across internal research, assets, programs and external evidence. ### Therapeutic Area Core Structured knowledge for a disease area, modality or biological domain. ### Asset Core A continuously updated view of a target, molecule, program or portfolio. ### Regulatory Core Internal and external regulatory knowledge organized around products and markets. ### Customer Core A scientific intelligence capability embedded into a platform or customer-facing product. ### Content Core Controlled access to proprietary content through search, workflows and applications. Core engagement options ### Core Discovery - Use-case and stakeholder workshops - Data and systems assessment - Initial domain and workflow mapping - Architecture recommendation - Pilot and rollout plan What's included (5) - Use-case and stakeholder workshops - Data and systems assessment - Initial domain and workflow mapping - Architecture recommendation - Pilot and rollout plan Outcome: A \*Core specification and implementation roadmap. ### Core Pilot - A scoped set of internal and external sources - Initial ontology or entity model - One or more priority workflows - User interface or API access - Evaluation with representative users What's included (5) - A scoped set of internal and external sources - Initial ontology or entity model - One or more priority workflows - User interface or API access - Evaluation with representative users Outcome: A functioning \*Core pilot with documented performance and production requirements. ### Production Core - Production data integrations - Identity, access and governance - Custom workflows and applications - Continuous data updates - Monitoring and evaluation - Training and operational support - Joint roadmap What's included (7) - Production data integrations - Identity, access and governance - Custom workflows and applications - Continuous data updates - Monitoring and evaluation - Training and operational support - Joint roadmap Outcome: An organization-specific intelligence capability that you run, or run jointly with Amass. [Book a demo](https://amass.tech/demo) What the work is made of ## Services across the lifecycle ### Discovery & design - Executive and stakeholder workshops - Use-case identification and prioritization - Technical feasibility - Data and systems mapping - Success metrics - Security and deployment planning Services (6) - Executive and stakeholder workshops - Use-case identification and prioritization - Technical feasibility - Data and systems mapping - Success metrics - Security and deployment planning ### Training & enablement - User onboarding - Research workflow training - Developer and API training - Team-specific playbooks - Train-the-trainer programs - Recurring office hours Services (6) - User onboarding - Research workflow training - Developer and API training - Team-specific playbooks - Train-the-trainer programs - Recurring office hours ### Implementation - API integration - Internal data connection - SSO and access management - Workflow and agent development - Custom application development - Evaluation, testing and deployment Services (6) - API integration - Internal data connection - SSO and access management - Workflow and agent development - Custom application development - Evaluation, testing and deployment ### Operations & optimization - Monitoring and incident support - Usage and quality reviews - Workflow optimization - New data-source activation - Model and retrieval tuning - Roadmap reviews and ongoing engineering capacity Services (6) - Monitoring and incident support - Usage and quality reviews - Workflow optimization - New data-source activation - Model and retrieval tuning - Roadmap reviews and ongoing engineering capacity The delivery process ## How an engagement works Most engagements begin with a scoped pilot. If you already have a specification, you can start later. Each phase ends in named deliverables, the same ones your statement of work cites. 1. 01 ### Align Establish the business objective, users, systems, constraints and intended outcome. Use-case briefStakeholder mapInitial scope 2. 02 ### Design Map the workflow, data sources, technical architecture, evaluation criteria and deployment model. Solution designDelivery planSuccess metrics 3. 03 ### Prove Build a scoped prototype or pilot using representative workflows and data. Working pilotEvaluation resultsProduction recommendation 4. 04 ### Deploy Complete the integration, testing, security work, training and production rollout. Production implementationDocumentationHandover 5. 05 ### Operate Monitor, maintain and improve the deployment through an agreed support or forward-deployed model. Ongoing optimizationNew workflowsOperational reporting ### Who you work with Forward-deployed engineer Ships integrations, agents and pipelines in your repo and your environment. Scientific lead Owns evaluation frameworks and domain modeling with your researchers. Engagement lead One accountable owner for scope, phases and deliverables, from Align to Operate. Everything runs on enterprise cloud infrastructure in EU regions, through your IT and security review. [See how we handle security →](https://amass.tech/security) What this looks like in practice ## Example engagements Six shapes an engagement takes, each with what it delivers and the level it maps to. ### Internal intelligence workflow Connect Amass to internal scientific documents and deploy a governed workflow for target, indication or asset research. Governed workflow · weekly run 12 internal reports linked to 340 external sources 3 new Phase II readouts flagged for review Every claim cited: PMID or DOI, trial ID or internal doc ID Delivered as: ImplementAlign → Prove → Deploy ### Customer-facing scientific intelligence Embed Amass into a CRO, CDMO, software platform or marketplace to provide evidence-backed intelligence directly to customers. » GET /cores/trialcore/records?query=NASH&phase=PHASE3&limit=10 » 200 · 10 records » "nctId": "NCT03900429", … Delivered as: Implement → Forward DeployAlign → Design → Deploy → Operate ### Custom scientific agent Design, evaluate and deploy an agent for a defined workflow such as asset diligence, regulatory monitoring or clinical landscape analysis. 1Scan 4 cores for KRAS-G12C landscape 2Rank 8 candidates by trial status 3Draft diligence brief · 47 citations Delivered as: Forward DeployDesign → Prove → Operate ### API implementation and migration Move from an existing search or data provider to the Amass API, including mapping, integration, testing and rollout. » map: legacy\_id → amass record id » coverage check: 98.4% resolved » cutover: staged, per workflow Delivered as: ImplementAlign → Design → Deploy ### Organization-wide enablement Train scientific, commercial and technical teams, establish reusable workflows and support adoption through recurring office hours. Adoption plan · quarter one 4 role tracks · 6 team playbooks Prioritized use-case backlog (11) Office hours · biweekly Delivered as: EnableAlign → Deploy ### Custom \*Core Build a persistent intelligence layer around a company, therapeutic area, portfolio, proprietary corpus or customer-facing product. DomainTherapeutic-area \*CoreSourcesInternal corpus + 3 cores + licensed dataGovernanceSSO, access tiers, full traceability Delivered as: Forward DeployAlign → Design → Prove → Deploy → Operate After the build ## Support designed around your deployment Each engagement level pairs with a support model of the same name, so there is no second menu to learn. Standard adoption with Enable Typically available with team and enterprise subscriptions. - Structured onboarding - Standard technical support - Documentation and implementation guidance - Initial team training - Periodic adoption reviews Included (5) - Structured onboarding - Standard technical support - Documentation and implementation guidance - Initial team training - Periodic adoption reviews Guided implementation with Implement For customers deploying a defined integration or workflow. - Named implementation lead - Discovery and architecture sessions - Configured implementation plan - Developer support - Testing and launch support - Team training and handover Included (6) - Named implementation lead - Discovery and architecture sessions - Configured implementation plan - Developer support - Testing and launch support - Team training and handover Strategic deployment with Forward Deploy For larger, multi-team or business-critical programs. - Dedicated engagement lead - Forward-deployed engineering capacity - Custom development and integration - Joint roadmap and governance - Ongoing production support - Recurring scientific and technical reviews Included (6) - Dedicated engagement lead - Forward-deployed engineering capacity - Custom development and integration - Joint roadmap and governance - Ongoing production support - Recurring scientific and technical reviews The order form or statement of work sets the service level, based on scope, complexity and how much ongoing ownership you want from us. ## Start with the outcome Tell us what your team needs to deliver, where the workflow should live and how much help you want. We’ll recommend where to start: enablement, a scoped implementation, a forward-deployed engagement or a custom \*Core. [Book a demo](https://amass.tech/demo)[Try free](https://amass.tech/try) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Amass for SciSure — The science behind your experiments Amass for SciSure # Explore the science behind your experiments Amass and SciSure are bringing scientific intelligence directly into the lab workflow. Add your experiments to a research question and follow the evidence across literature, trials, patents, drugs and genes, without leaving SciSure. [Get started with SciSure](#get-started)[See how it works](#how-it-works) Choose the experiment sections you share. Inspect the sources behind the response. Runs inside [Jump to an example response](#example-response)[View full size (opens in a new tab)](https://amass.tech/scisure/01-amass-inside-scisure.png) ## Scientific intelligence inside your lab notebook Amass connects scientific sources through specialized \*Cores, so you can follow one question across domains in one conversation. The SciSure integration adds your own experiments as context. Researchers organize and document their experiments in SciSure's electronic laboratory notebook. Amass runs inside it. - Literature - Clinical trials - Patents - Regulatory - Drugs - Genes - Live web How it works ## From your experiment to the supporting evidence 1. Step 1 ### Bring your experiments into the conversation Search your SciSure experiments and add up to five experiments that bear on your question. Amass then reads the evidence in the context of your work. [View full size (opens in a new tab)](https://amass.tech/scisure/02-select-experiments.png) 2. Step 2 ### Choose what Amass reads Preview the text, select sections and optionally mask compound IDs before the context joins your conversation. [View full size (opens in a new tab)](https://amass.tech/scisure/03-review-shared-context.png) 3. Step 3 ### Ask about your own research Ask Amass to review a methodology, investigate possible explanations for an unexpected result or find evidence that challenges a hypothesis. Your experiment is the starting point. [View full size (opens in a new tab)](https://amass.tech/scisure/04-ask-about-experiment.png) 4. Step 4 ### Review the answer and its sources Read the suggestions next to the evidence behind them. The references separate sources cited from sources reviewed, and follow-up questions stay inside SciSure. [View full size (opens in a new tab)](https://amass.tech/scisure/05-answer-and-references.png) Example workflow using demonstration experiment data. Example questions ## Questions to ask about your experiment - Review this methodology against relevant published studies. Where are the gaps? - What could explain this unexpected result, and which evidence supports each explanation? - Find evidence that challenges our working hypothesis. - Which follow-up experiments could help distinguish between these explanations? Open the references and judge how the evidence applies to your study before you act on the answer. Your data ## You choose the context you share Select the experiment sections relevant to your question and review the text before you share it with Amass. The integration uses selected text; experiment attachments and images are excluded. ### Preview the content See selected experiments, sections and an approximate word count. ### Mask compound IDs Optionally replace compound identifiers with masked labels. Names, sample codes and patient identifiers are not masked by this control. ### Check the scope The interface shows which sections are included and whether compound IDs are masked. Customer data, including queries, documents, results and outputs, is never used to train, fine-tune, or improve AI models. This applies to Amass models and to any third-party model providers we use. [Read about Amass data security](https://amass.tech/security) Get started ## Get started with Amass in SciSure Open Amass from your SciSure workspace and start with a research question. ### Already using SciSure? You need an Amass account and access to the integration in your SciSure workspace. Once it's enabled, open the Amass tab and ask a question, or add an experiment for context. [Create an Amass account](https://app.amass.tech) Already have an account? Sign in when prompted in SciSure. [Open the SciSure marketplace listing](https://www.scisure.com/marketplace/amass) Need help getting connected? [Contact Amass](mailto:hj@amass.tech?subject=Amass%20for%20SciSure) ### Exploring the integration for your team? See how your team could use experiment context and external scientific evidence together. [Book a SciSure integration demo](https://calendar.app.google/kBpdDJrS7h9bfqjHA) FAQ ## Questions about Amass in SciSure Can I ask questions without attaching an experiment? Yes. Start with a scientific question, or select experiments to add context from your own work. What can I add from SciSure? Select up to five experiments and choose the sections to include. The integration shares text only; attachments and images are excluded. The selector lets you browse the latest 100 experiments. Can I check the evidence behind a response? Yes. Inspect the references and distinguish sources cited in the response from those reviewed during the research. What does compound ID masking cover? It replaces compound identifiers with masked labels. Names, sample codes and patient identifiers are not masked by this control. Review the selected text before sharing it. What does access cost? Contact Amass to confirm access and pricing for your SciSure workspace. General Amass plans are available on [our pricing page](https://amass.tech/pricing). ## Bring your next research question to SciSure Add your experiments as context and follow the evidence with Amass. [Get started with SciSure](#get-started)[Book a demo](https://calendar.app.google/kBpdDJrS7h9bfqjHA) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Security Security # Your data stays private, and it stays yours. You work in a regulated environment and need to know where your data goes. Amass keeps each customer’s index separate, hosts it in EU regions and never trains models on your data. EU hostingEnterprise cloud infrastructure Infrastructure ## Cloud infrastructure hosted in EU regions The Amass stack is hosted on enterprise cloud infrastructure in EU regions. ### Managed Kubernetes Container orchestration with provider SLAs, private networking and automatic failover. ### Model providers under contract Our third-party model providers are contractually barred from retaining or training on your data. Enterprise data processing terms apply by default. ### Hardware-backed secrets + managed PostgreSQL Secrets in hardware-backed key storage, and a managed relational database encrypted at rest and in transit. Single-tenant and bring-your-own-cloud deployments are available if you need stricter isolation. Data privacy ## Your queries never touch a training pipeline. Customer data (queries, documents, results and outputs) is never used to train, fine-tune or improve AI models, ours or those of the third-party providers we use. The commitment is written into our contracts. Retrieval and generation are separate. Private data is retrieved from your authorized index and used only inside the isolated context of that request. It never feeds public web search, cross-customer learning or model training. - Tenant isolation: your data index is never co-mingled with other customers - No public indexing: your documents are never surfaced to other users or search engines - Right to deletion: request log deletion at any time - Usage logs: for operational monitoring only, never analytics or training Access control ## Identity and permissions through your IdP Amass connects to your identity provider and enforces role-based access per workspace and per tenant. ### SSO SAML, OIDC, SCIM via Microsoft Entra ID or your IdP ### RBAC Role-based access control per workspace ### MFA Multi-factor authentication required on every account ### Just-in-time provisioning Users are provisioned and deprovisioned automatically ### Per-tenant isolation No cross-tenant data access Certifications & compliance ## Controls in place, certifications in progress. Your security team can review our controls, architecture and policies now, without waiting for a certificate. ### ISO 27001 ~6 months Controls fully mapped. Formal certification in progress (est. ~6 months). Right-to-audit provisions available now. ### SOC 2 Type II ~12 months Controls fully mapped. Formal certification in progress (est. ~12 months). Security questionnaires answered under NDA. ### NIS2-aligned Active Governance and incident response aligned to NIS2 requirements. EU-hosted infrastructure supports national regulatory requirements. If your infosec team needs more detail, we share architecture documentation, security questionnaires and right-to-audit provisions under NDA. [Request security documentation](mailto:support@amass.tech) Infrastructure partners ## Enterprise terms with every provider Every third-party infrastructure provider works under a data processing agreement. - Microsoft Azure - Azure OpenAI - Azure Kubernetes Service - Azure Key Vault ## Have questions about security? We answer infosec questionnaires and walk your security team through the architecture, controls and policies. [Talk to us →](mailto:support@amass.tech)[Book a security review →](mailto:support@amass.tech?subject=Security%20Review%20Request) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # HACK_002: Signal From Tomorrow, Vienna HACK\_002: Signal From Tomorrow, HOIV Vienna, 26–27 September 2026 # Three science tracks, and twenty-four hours to build one of them. Amass is a partner of [HACK\_002](https://events.teloscircle.com/hackathon-02/), the hackathon running under the line “building the future we were promised”, and every team gets **$500 in Amass API credits**. Amass is one REST API over six cross-linked \*Cores: 43M+ papers, 1.2M+ clinical trials, 22K+ molecules, 43K+ genes, FDA and EMA authorizations, and patents. Every answer arrives with the record it came from. You do not need to be a developer to use any of it, and you do not need a biology degree to enter Block B. Each of the three tracks below opens with a picture of what a finished entry looks like, then three concrete places to start, then a prompt you can paste straight into Cursor. [Get started →](https://amass.tech/onboarding/signal-from-tomorrow)[Pick your track ↓](#tracks) Doors open at 12:00 on Saturday, building runs from 13:00 to 13:00, and judging follows lunch on Sunday. The venue is HOIV, Arsenalstraße 11, 1030 Vienna, with 24-hour access. Registration and the full schedule are on [Luma](https://luma.com/1ka4b1nm) and the [Telos Circle event page](https://events.teloscircle.com/hackathon-02/). Questions for us: [hello@amass.tech](mailto:hello@amass.tech). Pick a track ## Which one is yours Choosing is the first and largest decision of the weekend, so here are the three side by side. Nothing stops you from moving later, but the entries that land tend to be the ones that picked at 13:00 on Saturday and never looked up. B1 ### AI for Longevity & Aging Choose this if You like working with real data, you can read a statistics table, or you already have opinions about ageing biology. You will build A tool, model, analysis, or hypothesis about ageing and healthspan. Judged on Held-out error, or an effect estimate with its uncertainty. Explain the sampling and the confounding, and say why the result does not establish a treatment benefit. [See the starting points](#longevity) B2 ### Neural Systems Choose this if You like dynamical systems, graphs, or simulation, and you would rather watch something behave than fit something to a table. You will build A simulation or an experiment around neural and brain-inspired systems. Judged on Pattern recall, response delay, network stability, or another metric you define, measured across a parameter sweep and against a baseline. [See the starting points](#neural) B3 ### AI for Scientific Discovery Choose this if You like building tools and agents, and you are more interested in how research gets done than in any single result. You will build A system that speeds up one stage of the scientific process. Judged on Citation support, reviewer agreement, false positives, or time saved, against a manual or simple baseline. A generated hypothesis is not a validated discovery, and saying so in the demo counts in your favour. [See the starting points](#discovery) A note on judging Block B is judged by **Constantin Convalexius** (Medical University of Vienna, Harvard) and **Lucas Mair** (Technical University of Munich, ProviGen), alongside the general panel. They will have seen the underlying literature before. The Measure line on each card above is the closest thing to a published rubric, so treat it as one. [B1 Longevity & aging](#longevity)[B2 Neural systems](#neural)[B3 Scientific discovery](#discovery) Track B1 ### AI for Longevity & Aging Biomarkers, ageing clocks, biological data, literature, and the research workflows around them. Public datasets are plentiful here, so this is the track with the shortest distance between sitting down and having a first result on screen. BioMedCoreGeneCorePublic datasets How it is measured Held-out error, or an effect estimate with its uncertainty. Explain the sampling and the confounding, and say why the result does not establish a treatment benefit. localhost:3000/evidence-map #### Which ageing clock says what 4 studies · 5 clocks · JuFo 2+ Four published studies down the side and five DNA methylation clocks across the top. Each cell reports whether that study found significant epigenetic age acceleration, significant deceleration, a non-significant result, or did not report the clock at all. | Finding | Horvath | Hannum | PhenoAge | GrimAge | DunedinPACE | | --- | --- | --- | --- | --- | --- | | IPF lung tissue vs healthy controlsacceleration tracks disease severityAm J Physiol Lung Cell Mol Physiol 2025JuFo 2PMID 39970931 | ↑ | ↑ | ↑ | — | ↑ | | Fast vs slow glaucoma progressionHorvath +2.93 y, Hannum +1.24 yOphthalmology 2025JuFo 3PMID 39716635 | ↑ | ↑ | ns | ↑ | — | | Blood vs brain-MRI age, twin cohortheld up in within-pair modellingBrain 2025JuFo 3PMID 39992867 | ns | ↑ | ns | ↑ | ns | | Drug-naive first-episode schizophreniadeceleration, p = 0.01BMC Psychiatry 2023JuFo 2PMID 36650462 | ↓ | ns | ns | — | — | ↑ accelerated↓ deceleratedns not significant— not reported Read the Horvath column downwards. It runs up, up, ns, down across four studies, so the clock you pick is a choice you have to defend. Three places to start #### Age-stratified biomarker panel Pick three biomarkers and plot them by age decade in NHANES. Write down every exclusion you made and say plainly what the sampling cannot tell you. The write-down is the hard part and the part that scores. NHANES · half a day #### One signature, two studies Take a single ageing-related expression signature and compare it across two GEO studies. Then find the batch effect and show it, before a judge asks whether you looked. GEO · one evening #### Evidence map of a mechanism Link a proposed ageing mechanism to the papers behind it and mark the places where they contradict each other. This is the one in the window above, and it is the least code of the three. BioMedCore · a few hours Paste this into Cursor to start prompt ``` Build me an evidence map for one ageing question. Question: does epigenetic age acceleration track disease severity, and do the different clocks agree with each other? Use the Amass API at api.amass.tech with my key in AMASS_API_KEY. Search BioMedCore for studies since 2023, keep only journals with journalQualityJufo 2 or higher, and for each paper pull out which clocks were measured and which direction each one moved. Lay it out as a matrix with one row per study and one column per clock, put the PMID on every row, and mark every cell where two studies disagree. Do not average the disagreements away. ``` The call behind the picture bash ``` curl "https://api.amass.tech/api/v1/cores/biomedcore/records\ ?query=epigenetic+clock+DNA+methylation+biological+age+acceleration\ &minJournalQualityJufo=2&minPublicationDate=2023-01-01&limit=6" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · 6 records, JuFo ≥ 2 ``` TITLE JOURNAL DATE JUFO CITED Epigenetic age acceleration in idiopathic pulmonary… Am J Physiol Lung 2025-03-01 2 10 Accelerated Epigenetic Aging Is Associated with Fast… Ophthalmology 2025-05-01 3 9 Epigenetic age acceleration in peripheral blood corr… Brain 2025-08-01 3 5 Extracellular vesicles and epigenetic aging clocks i… J Gerontol A 2026-08-04 3 0 Low birthweight is associated with epigenetic age ac… Evol Med Public Hlth 2023-01-01 2 15 Epigenetic clock analysis of blood samples in drug-n… BMC Psychiatry 2023-01-17 2 13 ``` Starter data for this track [ #### CDC NHANES Population health and biomarker survey data, free to download. Open](https://wwwn.cdc.gov/nchs/nhanes/)[ #### NCBI GEO DataSets Public gene-expression studies, including most ageing signatures. Open](https://www.ncbi.nlm.nih.gov/gds) Also try GeneCore turns an ageing target into a defensible pick in one call. SIRT1, FOXO3, or whatever your hypothesis rests on comes back with its tractability, its genetic constraint, and its curated safety signals, each with the number behind it. That is the paragraph in your write-up that says why this gene and not another one. Track B2 ### Neural Systems Neural simulation, in-silico nervous systems, connectomics, and emergent behaviour. This is the track where a small, complete thing beats a large, half-wired thing: a hundred neurons that demonstrably learn one pattern is a finished entry. Brian2Allen Brain MapGeneCore How it is measured Pattern recall, response delay, network stability, or another metric you define, measured across a parameter sweep and against a baseline. localhost:8888/network #### Model card · the assumptions in your network SCN1A · Nav1.1 LOEUF 0.098decile 0 pLI 1.00loss intolerant Dependent lines 0 / 1,244not essential PDB entries 7DTD3D solved What it is Nav1.1, the pore-forming alpha subunit of a voltage-gated sodium channel. 2,009 aa, 229 kDa, four homologous domains of six transmembrane regions each. Kinetics you can set Phosphorylation at Ser-1516 by PKC slows inactivation and reduces peak sodium current. That is a knob in your model with a measurement behind it. What breaking it does Variants cause Dravet syndrome, generalized epilepsy with febrile seizures plus, and familial hemiplegic migraine 3. Your perturbation has a clinical phenotype attached. ENSG00000144285UniProt P354982q24.3 Your network, trained Weights shuffled The two panels are yours to produce. The card above them is the part you do not have to guess at, and it is the part a judge can check. Three places to start #### A hundred neurons that learn one pattern Train a small spiking network on a single pattern, then shuffle the synaptic weights and run it again. The comparison against the shuffled control is what turns a demo into a result. Brian2 · one evening #### Motifs under edge removal Take a small connectivity graph, delete a chosen set of edges, and show how the motif statistics move. Say which edges you chose and why before you show the plot. Allen Brain Map · half a day #### A sensory-to-motor loop that degrades Wire a simulated loop from sensing to movement, then raise the noise and the signal delay and plot where the response stops being stable. Name the breaking point. simulation · a few hours Paste this into Cursor to start prompt ``` Help me build a small spiking network and a real baseline. Goal: a 100-neuron spiking network in Brian2 that learns one input pattern, plus a control with the synaptic weights shuffled. Set it up so I can sweep one parameter and plot pattern recall against that sweep, with the control on the same axes. Give me a spike raster for both conditions. Then write the assumptions section for me: list every biological parameter I have hard-coded, and mark which ones I picked out of convenience rather than from a measurement. ``` The call behind the picture bash ``` curl "https://api.amass.tech/api/v1/cores/genecore/records\ ?query=SCN1A&include=protein&limit=1" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · GeneCore SCN1A ``` symbol SCN1A ENSG00000144285 2q24.3 2,009 aa 229 kDa targetClass Ion channel › Voltage-gated ion channel › Voltage-gated sodium channel constraint LOEUF 0.098 pLI 1.00 (LOEUF decile 0) missense o/e 0.564 essentiality not essential, 0 dependent lines of 1,244 screened structure 3D: yes · PDB 7DTD · Pfam PF00520, PF24609, PF06512, PF11933 ptm Ser-1516 phosphorylation by PKC slows inactivation, lowers peak current disease Dravet syndrome · GEFSP2 · familial hemiplegic migraine 3 ``` Starter data for this track [ #### Allen Brain Map Atlases and connectivity data across species. Open](https://brain-map.org/)[ #### Brian2 tutorials The spiking-network simulator, from a standing start. Open](https://brian2.readthedocs.io/en/stable/resources/tutorials/index.html)[ #### fly-behavior-experiments A repository you can fork rather than start from an empty file. Open](https://github.com/The-Uploading-Lab/fly-behavior-experiments) Also try Amass does not simulate anything, and it is not meant to on this track. What it does is settle the assumptions argument. One GeneCore call tells you what is actually measured about the channel or receptor your model leans on, so the assumptions slide cites a record instead of a memory. Track B3 ### AI for Scientific Discovery Knowledge synthesis, scientific agents, experiment design, lab automation, and automated analysis. The trap on this track is building something that sounds right; the entries that land are the ones that show their sources and show where they failed. All six \*CoresCross-Core linksMCP How it is measured Citation support, reviewer agreement, false positives, or time saved, against a manual or simple baseline. A generated hypothesis is not a validated discovery, and saying so in the demo counts in your favour. localhost:3000/review-queue #### Claim check, awaiting a human 4 records · 2 flags Claim under test “Lecanemab is approved for early Alzheimer’s disease in the United States and the European Union on the same terms.” - ✓RegulatoryCoreLeqembi · FDAAMRC\_PxJ6… BLA · active · authorized 2023-01-06 · Eisai Inc · initiate at MCI or mild dementia - ⚑RegulatoryCoreLeqembi · EMAAMRC\_Nvtg… centralised · active · authorized 2025-04-15 · Eisai GmbH The EU indication is restricted to ApoE e4 non-carriers and heterozygotes. The US one is not. The claim of identical terms fails here. - ⚑TrialCoreClarity AD · NCT03887455AMTC\_F4vF… Phase 3 · Eisai · N = 1,906 · active, not recruiting · completion 2029-06-30 hasResults is false. The 18-month outcome everyone quotes is in the literature, not in this registry record. A tool that reads only the registry reports nothing here. - ✓TrialCoreStudy 201 · NCT01767311AMTC\_Jux8… Phase 2 · N = 856 · completed 2024-12-10 · results posted Verdict: partly supported. Two rows need a person.nothing ships until somebody signs off The value is not the answer. It is that every line carries the record it came from, and that the tool says where it is not sure. Three places to start #### An evidence graph for one question Answer a single research question with traceable papers, and make the tool state out loud which claims it cannot support. The refusals are the interesting output, not the answer. BioMedCore · one evening #### An experiment planner Propose the controls and a cost-aware design for one real lab question, then put it in front of somebody who runs that experiment and record what they changed. half a day #### A QC notebook that logs its decisions Find the outliers in a public assay table and write down every flagged row with the reason it was flagged. A reviewer should be able to overturn any one of them. a few hours Paste this into Cursor to start prompt ``` Build me a claim checker with a review step. Input: a claim about a drug, like "lecanemab is approved for early Alzheimer's disease in the US and the EU on the same terms". Use the Amass API at api.amass.tech with my key in AMASS_API_KEY. Search RegulatoryCore for every FDA and EMA authorization of that substance, and TrialCore for the trials behind it. For each record, say whether it supports or contradicts the claim, and quote the specific wording that decides it. Flag anything you are not sure about instead of resolving it, and refuse to output a verdict until a human has cleared the flags. Show me the failures as prominently as the passes. ``` The call behind the picture bash ``` curl "https://api.amass.tech/api/v1/cores/regulatorycore/records\ ?query=lecanemab&limit=4" \ -H "Authorization: Bearer amass_YOUR_KEY" ``` response · 200 OK · FDA + EMA, unified ``` NAME AGENCY STATUS AUTHORIZED HOLDER SUBSTANCE Leqembi FDA ACTIVE 2023-01-06 Eisai Inc Lecanemab-irmb Leqembi Iqlik FDA ACTIVE 2025-08-29 Eisai Inc Lecanemab-irmb Leqembi EMA ACTIVE 2025-04-15 Eisai GmbH Lecanemab Kisunla EMA ACTIVE 2025-09-24 Eli Lilly Nederland B.V. Donanemab # the EMA indications restrict to ApoE e4 non-carriers and heterozygotes. # the FDA one does not. that difference is the whole finding. ``` Starter data for this track [ #### OpenAlex API The open scholarly graph, for coverage across every field. Open](https://docs.openalex.org/)[ #### NCBI E-utilities Raw PubMed access, if you want to build the plumbing yourself. Open](https://www.ncbi.nlm.nih.gov/books/NBK25501/)[ #### Open Targets API Target-disease evidence, as a GraphQL endpoint. Open](https://platform-docs.opentargets.org/data-access/graphql-api) Also try Cross-Core links are the shortcut on this track. A gene resolves to the drugs that target it, those to the trials that ran them, those to the papers behind them and the FDA and EMA labels that cover them. Follow the reference fields one call at a time and every hop in your graph arrives with its own id. Treat an empty reference array as “none recorded”, not as “no evidence”. The offer ## Claim your team’s $500 in API credits One code per team, worth **$500**. Sign up with the email you registered for the hackathon with, then ask us for the code. What $500 buys Roughly **10,000 searches** or **50,000 record fetches**. Search costs $0.05 per 20 results; get-by-id and lookup are $0.01 each. That is far more than a team gets through in 24 hours, so query freely and cache what you re-run. Start here ## Four ways in, one key All four reach the same data with the same key. If you do not know which one you want, start with the assistant and let it write the first call with you. Start with this ### The Amass build assistant A chat grounded in the Amass documentation, built to help you write against the API. Describe what you are trying to do, such as finding every completed trial of an amyloid antibody with posted results and the papers behind them. It answers with the endpoint, the filters, and the field names, so you are not guessing at parameter names from a schema you have not read. [Open the assistant →](https://platform.amass.tech/assistant) [ ### The platform Your account, credits, and API keys, plus the full documentation, the interactive API reference, and the app gallery. Everything the assistant reads from lives here. platform.amass.tech](https://platform.amass.tech)[ ### The Amass app The product itself. Ask a research question and get an answer with the records attached. Use it on Saturday morning to see what the data can answer before you write a line of code. preview.amass.tech](https://preview.amass.tech)[ ### MCP Connect Amass to Claude, Cursor, or Codex and query all six \*Cores in natural language while you build. It also gives you a judge-friendly demo with nothing to deploy. Connect the MCP](https://amass.tech/mcp) The data ## Six \*Cores, one API Same base URL, same auth header, same error shape across all of them. A paper, a trial, a molecule, a gene, an authorization, and a patent can resolve to the same entity, so you follow one thread instead of joining three exports by hand. ### BioMedCore 43M+ PubMed, PMC and conference abstracts /cores/biomedcore/records ### TrialCore 1.2M+ Clinical trials indexed /cores/trialcore/records ### DrugCore 22K+ ChEMBL drugs & molecules indexed /cores/drugcore/records ### RegulatoryCore 7K+ FDA & EMA documents indexed /cores/regulatorycore/records ### GeneCore 43K+ Human genes indexed /cores/genecore/records ### PatentCore 170M+ Life-science patents indexed /cores/patentcore/records Apps to fork ## You do not have to start from an empty file Every app in the gallery runs on the same API, the same key, and the same cited records as the tracks above. All of them were built with [Lovable](https://amass.tech/life-science-api), so there is a route through this weekend that involves no local setup at all. Fork one, point it at your key, and change the question it asks. [Browse the app gallery →](https://platform.amass.tech/app-gallery)[No-code recipes](https://amass.tech/life-science-api) platform.amass.tech/app-gallery Research #### From Gene to Approval One gene → every drug, trial, and FDA/EMA approval behind it. Research #### Graveyard & Garden 25 years of late-stage trials, halted or still alive. Research #### Clinical Trial Atlas Where a drug is tested, and which regions are enrolling. Research #### Indication Timeline A drug across its Phase-3 indications and approvals. Regulatory #### Transatlantic Divergence Desk FDA versus EMA outcomes, side by side. Clinical #### Pipeline Monitor A weekly digest of new Phase 2/3 trials by sponsor. Nineteen starter apps across research, regulatory, and clinical. Each one is a working answer to “what does a small app on this data look like?” When to meet ## How to reach us over the 24 hours We are not at the venue this time. We are on the event Discord for the whole weekend, and we run a mentor session on Saturday. Mentor session We run a **mentor session at 16:00**. Bring the question you are stuck on, even if the build is still a sketch. The rest of the weekend We are remote, and in [Discord](https://discord.gg/sEGaBHMhWa) for the full 24 hours, or at [hello@amass.tech](mailto:hello@amass.tech). Questions about credits, endpoints, or which \*Core holds what are all fair game. Playbook ## Six things worth knowing before 13:00 on Saturday These are what separate an entry that lands from one that spends the weekend fighting its own plumbing. 01 ### Narrow until it feels too small. Twenty-four hours is the real constraint, and every track’s brief opens by asking for one narrow, testable question. A finished small thing wins against an unfinished large one, and the judges can tell the difference from the back of the room. 02 ### Build the baseline first. Every Measure line in Block B asks for a comparator. Shuffled weights, a random classifier, the manual version of the workflow. The baseline takes twenty minutes and it is what turns your number into a result. 03 ### Show where it fails. Two of the three track briefs ask for failures explicitly. A demo that names its own failure mode reads as finished work. A demo that hides it reads as a demo that has not been tested. 04 ### Cite it or drop it. Every Amass record carries a stable id and a route back to its source, so there are no invented PMIDs and no invented NCT numbers. That is the whole reason to query the data rather than ask a model from memory. 05 ### Let the build assistant write the calls. Describe what you want in plain language and it answers with the endpoint, the filters, and the field names, straight from the documentation. That removes the part of the weekend where you guess at parameter names from a schema nobody has read. 06 ### Read the envelope, respect the limit. Results arrive under data and failures arrive under error, so check for an error before you parse. The limit is 60 requests per 60 seconds; on a 429, read Retry-After and back off. There is no paging, so narrow with filters instead. Prototype in Claude ## Sketch it in a chat before you write any app code. Connect the Amass MCP in Claude, Cursor, ChatGPT, or Codex and query all six \*Cores in natural language. Explore the data, find the filters that work, and only then drop the same queries into your build. On Sunday it doubles as the demo: the answers come back cited in the chat, with nothing to deploy. In Claude Code, install the Amass skill and let the model wire up your data layer for you. That is a short path from “what is even in here?” to a working prototype, which is the right question to answer on Saturday afternoon rather than Sunday morning. [Connect the MCP](https://amass.tech/mcp)Starter agent:[Python](https://github.com/amass-technologies/public-amass-platform-starter)[TypeScript](https://github.com/amass-technologies/public-amass-platform-starter-ts) ## Pick one, and go build it Claim your team’s $500, choose a track, and let the API do the reading. Narrow it until it feels too small, build the baseline, and show where it breaks. We would like to see it on Sunday. [Claim your team’s $500 →](https://platform.amass.tech)[Build assistant](https://platform.amass.tech/assistant)[Try the app](https://preview.amass.tech)[Connect the MCP](https://amass.tech/mcp) One code per team. Ask in the event Discord or at [hello@amass.tech](mailto:hello@amass.tech) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Skills & Prompts Skills & Prompts # Run Amass in any assistant Paste-ready prompts and packaged skills that turn Amass into finished analyst work in Claude, Codex, Copilot, or any MCP-compatible assistant. Each carries a real, validated answer and downloadable take-home files, and needs no build or deploy. [Connect Amass via MCP](https://amass.tech/mcp) Topic Mode Output 13 items [ pdf+1 196trials in scope competitive-intelligenceSkillFeatured ## Landscape Monitor Name a field; get it mapped across trials, papers, patents, drugs, genes and approvals into a landscape page with charts, a workbook and a one-page briefing, every record screened with its reason. Ask for an update later to see what is new and what changed. Open showcase ](https://amass.tech/skills/amass-landscape-monitor) [ xlsx 14trials watched | Short name | Source id | Registry | Copy of | | --- | --- | --- | --- | | Acute, 50/75 mg vs placebo | NCT04072354 | ClinicalTrials.gov | | | | ↳ EUCTR2019-000470-36-BG | EU CTR | NCT04072354 | | Acute, 75/100 mg vs placebo | NCT04092686 | ClinicalTrials.gov | | competitive-intelligenceSkill ### Watchlist Monitor Name the trials, papers, approvals or drugs you need to follow; get one Excel board that re-reads them through Amass at every check and tells you exactly what changed: status, dates, enrollment, results posted, label revisions, retractions, new links. ](https://amass.tech/skills/amass-watchlist-monitor)[ csv 12targets scored | kinase | kinase\_family\_chembl\_leaf | small\_molecule\_tractability | protac\_tractability | | --- | --- | --- | --- | | PLK1 | Other protein kinase PLK family | Advanced Clinical | predictive-only | | WEE1 | Other protein kinase WEE family | Advanced Clinical | predictive-only | | CDK1 | CMGC protein kinase CDK5 subfamily | Advanced Clinical | predictive-only | researchSkill ### Target prioritization matrix A general, interview-driven target-discovery workbench: name the biology and the modality you care about, and get back a scored table of candidate targets — druggability, essentiality, genetic constraint, curated safety, and the real number of drugs already targeting each — assembled from GeneCore intelligence and DrugCore/TrialCore/BiomedCore/PatentCore cross-links. ](https://amass.tech/skills/target-prioritization-matrix)[ csv 2approved, first-gen | record\_type | title | id | source | | --- | --- | --- | --- | | paper | Mechanisms of Resistance to KRASG12C Inhibitors | PMID:33466360 | Cancers (JuFo 2) | | paper | Oncogenic KRAS blockade therapy: renewed enthusiasm and persistent challenges | PMID:34607583 | Molecular Cancer (JuFo 3) | | paper | KRAS G12C Mutations in NSCLC: From Target to Resistance | PMID:34064232 | Cancers (JuFo 2) | competitive-intelligenceSkill ### Biomedical Evidence Scout Name a drug, target, mechanism, or indication; get a landscape briefing that maps the published literature AND the clinical pipeline together — every paper and trial carrying a real Amass ID, PMID, or NCT you can check. ](https://amass.tech/skills/amass-biomedical-evidence-scout)[ csv+1 7linked papers | rank | title | journal | publicationDate | | --- | --- | --- | --- | | 1 | Efficacy and Safety of NVX-CoV2373 in Adults in the United States and Mexico | The New England journal of medicine | 2022-02-10 | | 2 | Safety, Immunogenicity, and Efficacy of the NVX-CoV2373 COVID-19 Vaccine in Adolescents: A Randomized Clinical Trial | JAMA network open | 2023-04-03 | | 3 | Safety, Immunogenicity and Efficacy of NVX-CoV2373 in Adolescents in PREVENT-19: A Randomized, Phase 3 Trial | medRxiv (preprint) | 2022-09-21 | clinical-evidenceSkill ### Trial evidence trace Paste one clinical-trial ID; get the exact published papers that describe it — pulled from Amass's publication↔trial graph, with trust metadata and abstract-grounded relevance notes. ](https://amass.tech/skills/trial-evidence-trace)[ docx 5Phase-3 trials 1\. Pivotal Phase-3 program AMASS EVIDENCE BRIEF regulatoryPrompt ### Tirzepatide evidence brief Name a drug asset; get a citation-auditable Word evidence brief — pivotal Phase-3 trials, grounded efficacy and safety findings, and the trial↔paper links — every figure verbatim from Amass. ](https://amass.tech/skills/tirzepatide-evidence-brief)[ docx+1 6Phase-3 trials EXPLORER-HCM - NCT03470545 (16 describing papers) Submission Evidence Assembler - mavacamten (Camzyos) regulatorySkill ### Submission evidence assembler Name one drug asset; get a citation-auditable, Module-2.5-style evidence narrative — every pivotal Phase-3 trial, its describing papers, and a canonical Amass-ID audit trail — assembled from Amass's publication↔trial graph with trust metadata on every cited paper. ](https://amass.tech/skills/submission-evidence-assembler)[ pdf 2MAESTRO trials researchPrompt ### Resmetirom evidence snapshot Name a drug asset; get a one-page PDF evidence snapshot — regulatory status, pivotal trials, grounded efficacy figures and the trial↔paper link — every number verbatim from Amass. ](https://amass.tech/skills/resmetirom-evidence-snapshot)[ csv+1 10papers screened | PMID | AMBC | title | journal | | --- | --- | --- | --- | | 25037936 | AMBC\_LgGUu4qEoXy99ZBBdZt6kLYpaq9 | Modeling and research on research | Clinical chemistry | | 29534060 | AMBC\_VRlhiy7cmm9d5q13WjyMr2UA9I0 | Meta-research: Why research on research matters | PLoS biology | | 15878467 | AMBC\_Mk6C6o15EdZyjpf9Pl1vmAoKBoz | Early extreme contradictory estimates may appear in published research: the Proteus phenomenon in molecular genetics research and randomized trials | Journal of clinical epidemiology | research-integritySkill ### ORCID record self-verify Paste your own ORCID; get the publication record Amass indexes under that verified identity, screened in one pass for retractions, below-peer-review journals, and trial linkage — a clean integrity self-audit before a grant, tenure, or promotion submission. ](https://amass.tech/skills/orcid-record-self-verify)[ xlsx 10papers | KOL paper-trial triangulator — Mikhail Kosiborod · query "cardiometabolic heart failure" | | | | | --- | --- | --- | --- | | Top-10 by relevance (sample, not the full oeuvre). Source: Amass BiomedCore authorNames seed + per-paper referencesTrialCore edge. 6/10 trial-linked, 4/10 (meta-analyses/reviews) none. 0 retracted. | | | | | paper\_PMID | paper\_AMBC | title | journal | | 35228754 | AMBC\_7BiZMrt83VZHgdVwEesrmHqQ0MB | The SGLT2 inhibitor empagliflozin in patients hospitalized for acute heart failure: a multinational randomized trial (EMPULSE) | Nature medicine | researchSkill ### KOL paper-trial triangulator Name a KOL plus a context query; get their top-10 most relevant papers, each triangulated to the exact clinical trial it reports — pulled from Amass's publication↔trial graph, with trust metadata and a downloadable engagement-dossier workbook. ](https://amass.tech/skills/kol-paper-trial-triangulator)[ xlsx 18Phase-3 trials | NCT | acronym | drug (interventionNames) | sponsor | | --- | --- | --- | --- | | NCT01051219 | FELINE | Losartan | Newcastle-upon-Tyne Hospitals NHS Trust | | NCT03053050 | STELLAR-3 | SEL; Placebo to match SEL 6 mg; Placebo to match SEL 18 mg | Gilead Sciences | | NCT03900429 | MAESTRO-NASH | MGL-3196; Placebo; Liver Biopsy | Madrigal Pharmaceuticals, Inc. | competitive-intelligenceSkill ### Indication pipeline landscape Name one indication; get its current Phase-3 drug pipeline mapped across every sponsor — each program tagged novel-agent vs repurposed-generic and overlaid with a cross-core publication-evidence maturity count — in one downloadable Excel matrix. ](https://amass.tech/skills/indication-pipeline-landscape)[ csv+1 9claims checked Lecanemab is FDA-approved for early Alzheimer's disease. In its 18-month phase-3 trial it delayed cognitive and functional decline by ≈5 months versus… clinical-evidencePrompt ### Trust-filtered grounded answer Ask a clinical question; get an answer where every sentence is pinned to a high-quality, non-retracted paper — and anything the evidence can't support is refused, not fabricated. ](https://amass.tech/skills/trust-filtered-answer)[ csv+1 10retracted papers | rank | title | journal | year | | --- | --- | --- | --- | | 1 | A paradigm shift in cell-free approach: the emerging role of MSCs-derived exosomes in regenerative medicine | Journal of translational medicine | 2021 | | 2 | Intra-articular injection in the knee of adipose derived stromal cells (stromal vascular fraction) and platelet rich plasma for osteoarthritis | Journal of translational medicine | 2017 | | 3 | Glioma gene therapy using induced pluripotent stem cell derived neural stem cells | Molecular pharmaceutics | 2011 | research-integrityPrompt ### Retraction hotspot scan Name a research topic; get a checkable, downloadable list of the retracted papers Amass indexes on it — ranked by citation reach, with strict no-reason guardrails. ](https://amass.tech/skills/retraction-hotspot-scan) Get started ## Bring cited evidence into your assistant. Connect the Amass MCP once, then run any of these workflows on your own targets, with every answer traceable to source. [Connect Amass to Claude](https://amass.tech/mcp) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion --- # Try Amass free Try Amass # Try Amass free. Pick your way in: the app in your browser, the REST API in your code, or the MCP in the AI tools you already use. [ ## The Amass app No code. Ask a question, get a cited answer you can check. Try free ](https://app.amass.tech/api/auth/sign-up)[ ## The Amass API Typed JSON from six \*Cores, every record cited. Build research into your product. GET /api/v1/cores/trialcore/records?query=NCT04023552&limit=1 { "data": \[{ "nctId": "NCT04023552", "acronym": "Lp(a)HORIZON", "phase": "PHASE3" }\] } Get an API key ](https://platform.amass.tech/api-keys)[ ## Amass MCP A remote server and one sign-in. Claude, ChatGPT, Claude Code and Codex query Amass with citations. { "mcpServers": { "amass": { "type": "http", "url": "https://mcp.amass.tech/mcp" } } } Set up in two minutes ](https://amass.tech/mcp) Not sure which one? [Book a demo](https://amass.tech/demo) --- # Use cases Use cases # Research workflows for life science teams Validate an investment, scan competitor pipelines or build research tools, with Amass inside the workflow your team already runs. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) One answer in the Amass app: the trace of what it searched, the claims it cites and the references it read. By workflow ## What Amass helps you do Six research workflows, and the \*Cores each one draws on. ### Due diligence Validate a target, modality or company before capital goes in. Clinical data, patents and publications come back in one cited answer you can defend when the committee asks for the source. VC, Pharma BD, BiotechPatentCoreTrialCoreDrugCoreBioMedCore ### Competitive intelligence See what competitors are filing, testing and publishing in your targets and indications. Patents, trials, regulatory records and papers come back in one answer your whole team reads from the same sources. Pharma R&D, Biotech, VCTrialCorePatentCoreRegulatoryCoreBioMedCore ### Literature review Synthesize the evidence on a mechanism, target or indication from BioMedCore's 43M+ papers. Each finding links to the paper it came from, so you can check the passage before you rely on it. Pharma R&D, CRO, ConsultingBioMedCoreGeneCore ### Pipeline monitoring Follow trials, regulatory filings and publications in an indication over time. Ask what changed since your last look and get new registrations, status changes and approvals back with their sources. VC, Pharma R&D, BiotechTrialCoreRegulatoryCore ### Patent landscape and FTO Map patent families, find whitespace and flag freedom-to-operate risks for your compound or mechanism. PatentCore holds 170M+ patents from every major office worldwide, with claims, applicants and citation lineage. Pharma BD, Biotech, ConsultingPatentCoreTrialCoreDrugCore ### Market intelligence Combine approval history, patent filings and publication activity for market and regulatory context. KOL activity, competitor filings and sector trends each come with the citation your client will ask for. Consulting, VC, BiotechRegulatoryCoreBioMedCorePatentCore By audience ## Amass for your team Different teams ask different questions. Here's how each one uses Amass. ### Due diligence that holds up in committee Validate targets, map competitors and follow the fields your portfolio companies work in, with a citation behind each claim a partner might question. Pre-investment diligence Cross-reference published evidence, patent coverage and trial data on an asset before the term sheet. Portfolio monitoring Ask what changed in a portfolio company's targets and indications: new trials, status changes, competitor filings. Sector briefs Build a cited map of the competitors in an indication or modality, as a report you can share. PatentCoreTrialCoreDrugCoreBioMedCore [Book a demo](https://amass.tech/demo) > Amass gives our investment team one place to validate science, patents, and trials, with citations we can stand behind in committee. Luis Toledo Senior Scientist, Novo Holdings SeedLab Skills and prompts ## Start from a tested prompt Paste-ready prompts and skills for your assistant, each with a validated answer you can inspect. Featured [ 1 Indication pipeline landscape](https://amass.tech/skills/indication-pipeline-landscape)[ 2 Target prioritization matrix](https://amass.tech/skills/target-prioritization-matrix)[ 3 KOL paper-trial triangulator](https://amass.tech/skills/kol-paper-trial-triangulator) [Browse all skills and prompts](https://amass.tech/skills) ## Find the right workflow for your team Talk to us or start on your own. Every answer cites its sources across eight \*Cores, from 43M+ papers to 170M+ patents. [Try free](https://amass.tech/try)[Book a demo](https://amass.tech/demo) Data No training on your data Hosting EU-hosted, enterprise Isolation Tenant-isolated by architecture Certification ISO 27001, SOC 2, NIS2: in motion