I work on quantitative methodology at the intersection of causal inference and machine learning, with an emphasis on the design and analysis of experiments. My research focuses on causal machine learning, adaptive and algorithmic experimental design, and the evaluation of complex digital systems like social media platforms and AI.

Before returning to academia, I spent four years as a research scientist at Facebook Core Data Science, developing methods for adaptive experimentation in large-scale online systems. Much of what we learned there about managing sequential, goal-oriented experimentation is embedded in Ax, Meta's open-source adaptive experimentation platform. I was also part of the US 2020 Facebook and Instagram Election Study, a landmark collaboration between independent academics and Meta on the effects of social media during the 2020 US elections. Applied or methodological, most of my work comes back to a single question: how do we get the most information out of the experiments we run?