A little about me

I am a Research Scientist at Amazon, where I build statistical infrastructure for large-scale experimentation — designing estimators, implementing Bayesian methods, and translating causal inference theory into production systems that drive outcomes across the company.

I got here through nine years of applying statistical methods across academia, international development, and industry — building conflict prediction models at Princeton, forecasting food insecurity for the UN, stress-testing election forecasts at FiveThirtyEight, and developing open-source LLMs for disinformation detection at the UN Department of Peace Operations. I hold a Ph.D. in Political Science from Columbia University.

I’m driven by the belief that better statistical infrastructure leads to better decisions, whether that means more reliable experiments at a tech company, more credible forecasts in a humanitarian crisis, or more honest uncertainty in public-facing predictions. My research focuses on closing that gap: making rigorous inference accessible, scalable, and trustworthy in the systems where decisions actually get made.

My core areas include causal inference, Bayesian statistics, machine learning, experimentation design, and large-scale statistical computing.

The latest copy of my resume is here.