About

I studied data science at UC San Diego, with a minor in cognitive science, and graduated in 2021. I started at Intel as an intern in 2020 and have been there since.

Outside work I build things with coding agents, mostly to find out where they break. That has turned into a benchmark for LLM investing, a harness that scores agents on a deck-building game, a health dashboard agents can read but not change, and a few games.

What I've learned is not very glamorous. Agents do good work when the goal is written down, when they can check their own output, and when the things they are allowed to change are few and clearly named. Most of my time goes into those three.

At Intel

• Helped design an enterprise retrieval-augmented assistant that sends each question to SQL telemetry or to technical documentation, depending on which one can answer it.
• Built the tooling that checks the assistant: automated checks, batch tests, human review and prompt tuning, aimed at hallucinations and wrong tool choices.
• Wrote more than 30 reusable AI-assisted development workflows, each with a test harness and automated checks, for a team of about 20.
• Used cProfile and Polars to help make an internal data-processing and performance-analysis platform 12 times faster.
• Build and maintain pipelines that process several terabytes a week in Python, SQL and PySpark.