zeta-agent
Automating the discovery of automation tools.
About
Automating the discovery of automation tools.
Single-cell analysis in Python. Scales to >100M cells.
autoformal-fidelity — fidelity of a source→formal crossing in fibre terms: Agda theorems (Crossing.agda) showing passing round trips rule out collapse, round trips are blind to additions, and coverage yields bijection; semantic fidelity honestly declared undecidable, with pre-registered falsifiers.
YellowFruit third-party client
Bioinformatics web-platform visualisation dashboard creation
VERY experimental Nintendo Switch 2 emulator
A text-first starter kit for mathematical research with AI agents: policy, workflow, writing and citation guides, review skills, and LaTeX and Lean modules.
Pegasus Workflow Management System - Automate, recover, and debug scientific computations.
Historical Lean autoformalization harness with human review, Verso source views, dependency graphs, and machine audits.
Building agents that learn and plan from experience to achieve goals in big worlds, pursuing Oak Lab's mission and the Alberta Plan in Lean.
Galactic Dynamics in python
cython + htslib == fast VCF and BCF processing
ProofForge turns Meta's Muse models into provers of financial code: Muse Code with ProofForge as its MCP server repairs code against locked Lean 4 specifications, every verdict checked by six kernels and recorded in a Passport anyone can re-run.
Lean 4 (v4.34.1 / Mathlib) formalization of noncommutative information geometry, Peirce-Weyl units, Drazin-Moore-Penrose spectral splittings, Jordan-Lie operator algebras, and categorical inductive colimits.
Rust accelerator for operator manipulation for TensorCircuit-NG
Terras stopping-time density. Binomial tail and first descent. Not a proof that every orbit reaches 1. Copyright Benjamin Stanley Frohman.
BLAS Levels 1–3 for AWS Trainium via NKI (cuBLAS-equivalent) — GEMM with stationary-tile reuse, batched GEMM, TRSM, validated DF-MP2 for quantum chemistry.
Paper-driven biological research through critical reading, public omics reanalysis and testable hypothesis development across relevant biological contexts.
Read a research paper beside its Lean proof code, with links to exact statements and a diagram of how the proof fits together.
Agent harness for learning difficult things.