Tom Finzell

I'm a data scientist and astrophysicist based in the Twin Cities. I've spent my career using computation to crack hard problems — modeling radio emission from stellar explosions, building time-series algorithms to extract faint signals from noisy data, leading applied machine learning projects across a dozen different domains. Along the way, I became convinced that if you're going to build something, you owe it to the next person to build it right.

These days I'm building the Open Nova Catalog — a serverless data platform on AWS that ingests, validates, and serves astronomical observation data from public archives. I designed the whole thing solo: infrastructure as code, contract-first backend, event-driven ETL, interactive scientific visualizations. It's the project I'm most proud of, and it's a good window into how I think about engineering problems.

I'm currently looking for data science, ML engineering, or research roles in the Twin Cities where I can bring this same approach — scientific rigor, real engineering, and a bias toward building — to problems that matter.