Skip to content
View manu-tej's full-sized avatar
6 followers · 4 following
  • Hayward
  • 01:02 (UTC -07:00)

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx2Achievement: YOLOAchievement: QuickdrawAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx2Achievement: YOLOAchievement: QuickdrawAchievement: Arctic Code Vault Contributor

Highlights

  • Pro

Block or report manu-tej

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
manu-tej/README.md

Manu Tej Sharma Arrojwala

Bioinformatics scientist — 8 years applying ML and software engineering to biology, 6+ in oncology drug discovery at Arcus Biosciences (tumor-immune phenotyping, neoantigen & TCR workflows, multi-omics platforms and pipelines).

I build biology agents and the evaluations that test whether they run real analyses with calibrated uncertainty — systems that recognize when the evidence isn't there and stop, instead of returning a confident answer that's quietly wrong.

Selected work

  • coscientist — multi-agent hypothesis tournament with a truth-tracking eval harness grounded against GPQA-bio
  • ai-scientists — trust profile of biology agents on BiomniBench-DA (contamination, calibration, refusal, benchmark validity)
  • methods-graph — bioinformatics methods knowledge graph (EDAM + nf-core + BioContainers on Kùzu)
  • dogma — graph-grounded bioinformatics method guidance in the IDE

🔗 manu-tej.github.io

Pinned Loading

  1. ai-scientists ai-scientists Public

    Trust-profile benchmarking of biology AI agents on BiomniBench-DA: contamination, calibration, refusal, trajectory consistency, and benchmark validity.

    Python

  2. dogma dogma Public

    Dogma is an in-progress AI scientist platform for computational biology; this public repo contains its VS Code/Cursor and local-control prototype.

    Python

  3. methods-graph methods-graph Public

    Dogma's method-validity layer: a provenance-aware graph of bioinformatics methods, assumptions, diagnostics, containers, and data handoffs.

    Python