I build production-style AI systems — agentic RAG, a local multi-agent LLM platform, and a two-node multi-GPU inference cluster — not demos.
Status: Open to ML Infrastructure · MLOps · AI Application roles · Remote (US) · Available now
Selected work
An agentic retrieval system over a large enterprise technical-documentation corpus that measurably cut specification-research time — built self-initiated, now adopted by colleagues.
A modular, offline-capable multi-agent platform — OpenAI-compatible gateway, control plane, autonomy engine, RAG, and IDE integration — with a five-mode adversarial peer-review loop and zero hosted-LLM dependency.
A self-built two-node multi-GPU cluster serving local models with a privacy-first architecture, thermal-aware autonomy, and GPU-wedge diagnosis and recovery.
Background
I test flight-critical rocket hardware on the RS-25 program, where nothing ships until it is verified. On nights and weekends I pointed that same discipline at AI/ML infrastructure — self-taught and self-built.
The through-line is reliability engineering: measured-not-asserted numbers, config-as-truth, and fail-closed defaults. The habits that keep flight hardware safe are the ones I bring to trustworthy AI systems.
Toolkit