← research · example output · code checking backend…

Ask any researcher's papers — LabBrain

Name a researcher. LabBrain reads their actual papers — full text where open-access — and answers your question, grounded with DOI citations.

Who it's for: PhD applicants vetting an advisor, grant writers and collaboration scouts, researchers onboarding into a new field, or anyone doing a literature review on a specific person — where the unit is one author's corpus, not the whole field.

Researcher

Your question

Try: Hummer · free energy Melo · lipids Noé · ML + MD

The first question about a researcher builds their corpus — it resolves them on OpenAlex, ingests open-access full text, and indexes it (~30–60 s). After that it's cached and answers are fast.

Reading their papers…

Resolving the researcher on OpenAlex
Ingesting open-access full text (Europe PMC → PDF)
Building hybrid BM25 + dense index
Retrieving passages & reasoning with citations

elapsed 0s · first build can take up to a minute

⚠ The live LabBrain server isn't active right now. The web demo runs on an on-demand backend that may be offline. You can still:

See a full worked example: example Q&A output for Gerhard Hummer (summarize · ask · compare).
Run it yourself: python labbrain.py build "<researcher>" then ask "<question>" — github.com/gianyrox/labbrain.

What makes it different: other tools search all of the literature; LabBrain reads one researcher's library — assembled automatically from just a name — and answers from it, cited back to DOIs. The full tool also does summarize (a lab's research program) and compare (co-advisor / collaborator finder).
github.com/gianyrox/labbrain · example output · more research tools