I'm a Principal AI Engineer at Cribl, where I lead AI projects from research through production.

My work covers sensitive-data detection, incident investigation, coding agents, and data-pipeline tools. I've also built distributed software that runs across more than 250,000 edge nodes and moves petabytes of data each day. Previously, I worked on search, streaming, and machine learning systems at Splunk.

My open-source projects include AIKit, a unified TypeScript interface for OpenAI, Anthropic, and Gemini generation APIs, and Ra, an agent runtime with explicit tools, middleware, permissions, and budgets.

I worked with researchers at Carnegie Mellon and the USGS on soil-moisture modeling and forecasting how post-wildfire soils respond to rainfall. The research appeared at GECCO and IEEE DSAA. My patents cover machine learning in search pipelines, security analytics, generative AI, and sandboxing untrusted code.

Elsewhere: GitHub · LinkedIn · Google Scholar · X