Software engineer and lawyer
I spent a decade in legal services and access to justice. I represented families in housing, benefits, and custody cases. I also worked on making legal systems fairer and more efficient: I designed a volunteer lawyer program that served over 20,000 defendants in consumer debt court, and worked with jurisdictions across the U.S. to reduce jail populations as part of the MacArthur Foundation's Safety and Justice Challenge. Then I spent eight years engineering at Rakuten Advertising, from junior engineer to leading a squad. I built customer-facing and backend services, moved them onto Kubernetes on Google Cloud, built the CI/CD pipelines that ship them, and most recently built AI agents the operations team used to investigate production incidents.
Now I want to bring the two together, working on systems that have an outsized effect on people's lives and making them work better.
Recent projects
- intake-classification-eval: two model runs against a legal aid intake dataset from Stanford's Legal Design Lab, showing how much the prompt shapes the results.
- pad-lab: a small data platform on public FEC campaign-finance data, with Terraform, dbt, BigQuery, and a live dashboard.
- homelab: a Google Cloud lab with reusable Terraform modules and keyless CI.