I build reliable AI systems across agents, voice, evaluation, and full-stack product engineering. I like product work where the interesting part is making AI-generated behavior observable, verifiable, and safe to ship. Based in Seattle, I enjoy early-stage startup pace and ownership.
- agent workflows with explicit specs, provenance, and verification
- real-time voice systems, speaker attribution, and conversational interfaces
- evaluation harnesses that test AI behavior beyond demo cases
- full-stack products with Python, TypeScript, FastAPI, React, and PostgreSQL
- cloud infrastructure, observability, and failure recovery
- open-source tools for persistent context and human-AI collaboration
Currently exploring open-source tools for persistent context and human-AI collaboration.
Selected work: Provenance Guard · Speaker Attribution · EchoPersona



