0:whoami
Last login: Tue Oct 6 22:30:22 2026 on ttys000
AI Architect & Solution Engineer
Building agents that show their receipts — shipped & audited, not just demoed. Clinical AI on FHIR/SMART · Governed Memory · MCP · Python · TypeScript · JAVA · Rust · Claude Code · CoPilot
01Python1 stars
Deterministic AI-governance drift detection: evidence-backed findings, stated coverage gaps, human-approved publishing. Runs on WeaveMind Cloud (~$0.03/run); includes the first external evaluation of the Weft language.
02Python3 stars
Local RAG pipeline for medical Q&A over the Merck Manual: Mistral-7B via llama.cpp, ChromaDB, and MiniLM embeddings.
03TypeScript1 stars
Compass BlackBox IQ: flight recorder for autonomous agents. Git-backed agent memory, decision records, and a skill forge, exposed over MCP. Agents League @ AI Skills Fest 2026.
04Python1 stars
Hand-written Triton GPU kernels with honest benchmarks on a Jetson Orin Nano, plus a study track through LLVM, MLIR, Triton's pipeline, and TorchDynamo/Inductor
05TypeScript2 stars
Offline-first EEG neurofeedback prototype for the Mind Media NeXus-10: EDF/EDF+ verifier, montage configs, experiment scaffold, and a web dashboard.
06Python1 stars
Multi-model LLM Council (Karpathy-style 3-stage deliberation over OpenRouter), exposed as an MCP server for Claude Code.
made with forkfolio