I am a theorist with broad scientific training who loves to make beautiful things out of math and code.
I develop mathematical and conceptual foundations for fallible agents, which help us better understand AI systems of all kinds.
My work combines
machine learning, probabilistic graphical models, information theory, programming languages, category theory, microeconomics, causality, and logic.
See my research page for more information!
Currently, I am a postdoctoral fellow at l’Université de Montréal and Mila under Yoshua Bengio, and also work part-time (35%) as a senior machine learning research scientist at LawZero. I mentor for the MATS Program in both roles. I received my PhD in Computer Science from Cornell University in 2024, where I was advised by the great Joe Halpern. Before that, I did an MPhil in CS at the University of Cambridge under Mateja Jamnik. I earned three majors (Mathematics, Cell & Molecular Biology, and Computer Science) and three minors (Chemistry, Physics, Cognitive Science) as an undergraduate at the University of Utah, where I did research in pure math (tropical geometry, with Aaron Bertram) and applied machine learning (structured prediction for natural language, with Vivek Srikumar). Even before that, I made video games in my free time.
I am
married
to the one and only
Varsha Kishore.
I like to sing and improvise on the piano.