Joined April 2012
- Tweets 337
- Following 742
- Followers 594
- Likes 743
55 Photos and videos
Alexander Junge retweeted
One of the key priorities for launching our Amass MCP was to enable users use our high-quality, integrated, semantically enriched, and crosslinked *Core datasets in the AI tools they are already using.
1
1
1
40
Happy to see that our team is not only shipping new features but also contributing to making the open source tools we rely on better - in a human way.
In a world of breathless AI hype, especially in the coding space, its a good reminder why we work on software:
1
29
to make life for humans better and part of that is collaborating with others good at their craft around the world, in a human way. Of course, this is a small fix to an API docs tool we are using at Amass and AI is helping draft these fixes, but still, it's all about caring.
20
Alexander Junge retweeted
Build with the Amass API: live demos in Lovable, Claude Code, and Python โ and a Q&A. Rewatch the webinar now. ๐
Three live demos and highlights:
1) Vibe-coding an AI research assistant in Lovable in a single prompt
2) Plugging the Amass API into Claude Code as a skill
1
1
16
This is blog post by @softwaredoug gives a good overview of the agentic (scientific) search future we are building towards and are enabling for all our Amass API users: to build highly specialized retrieval systems and search tools, with the help of agents, easily.
1
1
1
51
Alexander Junge retweeted
Today we are launching Amass MCP ๐
You can now bring cited biomedical evidence directly into the assistants you already use: Claude, ChatGPT, Codex, Claude Code, and other MCP-compatible tools. Have you deployed Co-Pilot, we are also compatible.
1
1
33
Looking for *reliable* life sciences data directly inside the AI tools you are already using? We just launched the Amass MCP. ๐ Using your Amass account, access connected and enriched papers and trials directly inside Claude, ChatGPT, and other MCP-compatible clients.
1
2
228
Under the hood, it uses the same high-quality retrieval that powers our App and APIs: papers from BiomedCore, trials from TrialCore. Give it a try and let us know what you think.
11
๐ข๐ป ๐ช๐ฒ๐ฑ๐ป๐ฒ๐๐ฑ๐ฎ๐, ๐๐ผ๐ฟ๐ป ๐ฃ๐ฒ๐๐ฒ๐ฟ๐๐ฒ๐ป ๐ฎ๐ป๐ฑ ๐ ๐ฎ๐ฟ๐ฒ ๐ด๐ผ๐ถ๐ป๐ด ๐๐ผ ๐ฏ๐๐ถ๐น๐ฑ ๐ฎ ๐ฏ๐ถ๐ผ๐บ๐ฒ๐ฑ๐ถ๐ฐ๐ฎ๐น ๐ฟ๐ฒ๐๐ฒ๐ฎ๐ฟ๐ฐ๐ต ๐ฎ๐ด๐ฒ๐ป๐ ๐น๐ถ๐๐ฒ. ๐๐ฟ๐ผ๐บ ๐๐ฐ๐ฟ๐ฎ๐๐ฐ๐ต. ๐๐ป ๐๐ป๐ฑ๐ฒ๐ฟ ๐ฑ ๐บ๐ถ๐ป๐๐๐ฒ๐.
1
1
42
And if you've been ๐ฎ๐ด๐ฒ๐ป๐-๐ฐ๐๐ฟ๐ถ๐ผ๐๐ but haven't actually shipped one yet โ this is the lowest-friction way I know of to watch one come together end-to-end. No prior setup. No prerequisites. Just show up.
1
11
๐
May 6, 2:00 PM CEST. Free. 60 minutes. Live Q&A at the end.
Registration link: watch.getcontrast.io/registeโฆ
19
I am ๐ฟ๐ฒ๐ฎ๐น๐น๐ excited: you can now build on three years of hard-won biomedical retrieval methodology. Relevance ranking, journal quality signals, and trial-to-literature cross-references packaged into a single API endpoint with BioMedCore and TrialCore. More *Cores on the way
1
1
1
178
6) A quick demo of a vibecoded tool (in Lovable) comparing a PubMed and clinicaltrials.gov API-based agent to the same agent with the Amass Data API as a drop-in replacement: loom.com/share/1ab384e58d5a4โฆ
1
27
7) Our brand new website provides an API overview (amass.tech/api-access) and a blog post highlighting our retrieval capabilities (amass.tech/resources/why-bioโฆ).
29