David Bruns-Smith

I am a Stanford Data Science Postdoctoral Fellow, working with Guido Imbens. I work at the intersection of machine learning and macroeconomics. On the technical side, I study observational causal inference, especially in sequential decision makings settings. There are deep connections between causal inference, domain adaptation, and robust machine learning, and I have found this to be a very fruitful area of research. On the applied side, my work has focused on household income, consumption, and debt. A recent proliferation of high quality administrative data has made it newly possible to scrutinize macro-consumption models from a machine learning perspective. I'm especially interested in the potentially very rich interplay between theoretical economic models and causal machine learning.I finished my Ph.D. in Computer Science at UC Berkeley in 2024, advised by Avi Feller in Public Policy and Statistics, and Emi Nakamura in Economics. My thesis studied causal inference methodologies relevant for macroeconomic policy making. In a previous life, I built gene sequencing accelerators for FPGAs, developed compilers for custom architecture, and wrote parallel implementations of tensor decompositions.