About

I build AI agent systems across the application and platform layers.

I work on domain agents, prompt and tool design, workflow orchestration, runtime architecture, event contracts, retrieval infrastructure, evaluation loops, tracing, and delivery pipelines.

I want agents to be useful in real engineering environments. That means clear task boundaries, reliable tool use, observable execution, testable quality, reusable SDKs, and deployment paths that can survive production constraints.

I publish public work onGitHub.