Lenny Bronner

I’m a data scientist who works on elections, media, and the messy real-world data that both produce. Most of my career has been spent building statistical systems that have to work in public, under deadline, and often on incomplete data.

The work tends to fall into a few areas:

• Election data and modeling. Live forecasting from partial returns, polling averages, voter registration data, and estimating how different groups actually voted. I spent nearly a decade doing this at The Washington Post, where I built their first live election night model and later led the team that rebuilt it.
• Data infrastructure for newsrooms. Pipelines and tools that let journalists work with voter files, campaign finance filings, and other public data without needing a data scientist in the room. Some of that work contributed to reporting that won the 2022 Pulitzer Prize for Public Service.
• Personalization and online learning. Recommendation, metadata tagging, and production bandit systems for deciding what to show readers and when, including push notifications and recirculation.
• Explaining any of the above. I’ve written about our methods, presented them at conferences, and gone on live television to explain a forecast while it was running.

I have an M.Sc in Statistics and a B.Sc in Mathematical and Computational Science, both from Stanford.

My CV has the full detail — roles, methods, publications, and talks.

If you have a project you’d like to talk about, email me at lenny.bronner@gmail.com.