Simulation-first lab robotics: PyLabRobot arm control, eye-in-hand vision, dexterous manipulation, and auditable workcell tasks.
Capture, decode, and guarded-replay tooling for no-API lab instruments; incomplete maps remain blocked from actuation.
Closed-loop ML for biology: ESM-2 protein design by simulated annealing, Bayesian assay optimization with a split-conformal accept/reject/escalate gate, and an agentic design-build-test-learn engine driving a liquid handler. Sim-first, CPU-only.
Evidence and permission for autonomous labs: expert QC gates, workcell orchestration, provenance, recovery, and honest hardware boundaries.
A world model runs the bench. Track A entry for the AI for Science World Models Hack @ Zeon Systems: sim-first, composing the @di-omics autonomous-lab stack + a bridge to drive Zeon's arm via PyLabRobot.
Auditable MCP tools for PyLabRobot: run and sample context, typed provenance, simulation by default, and explicit hardware permission.
Find the cheapest sequencing library prep that still gives a callable answer. An agent maximizes unique reads per dollar, warm-started by a bench scientist and gated on in-process CV.
Three Claude Code skills for authoring PyLabRobot devices under the v1b1 capability-architecture: protocol extraction, scaffolding, and review.
Synthetic, ground-truth demos for scRNA-seq, methylation sequencing, PCR enrichment, and other omics QC; no wet-lab claims.
Generic, simulator-only scRNA-seq and whole-genome sequencing workflows, from planning through configurable analysis handoffs.
Add a description, image, and links to the pylabrobot topic page so that developers can more easily learn about it.
To associate your repository with the pylabrobot topic, visit your repo's landing page and select "manage topics."