Building production AI, scientific software, and data systems people can trust.
I build production AI, design scientific software, and turn complex workflows into clear, reliable data systems for scientific teams.
How I work
My work spans applied AI research and production data systems, scientific platforms, shared SDKs, and data assets. The through line is making complex capabilities understandable, trustworthy, and useful across teams.
Selected evidence
- Increased genomic variant mention recognition recall by 30 percentage points over an internal baseline through agent-based schema and prompt optimization research and development.
- Modernized scientific analysis pipelines and automated reporting across 50,000+ sequencing samples and 5+ TB of data.
- Authored and led development of shared SDKs and Dagster data assets used across departments.
- Led multidisciplinary teams delivering AI, scientific software, production APIs, and internal platforms.