I am an AI researcher and engineer. I believe in a first-principles approach to problem-solving and ground my work in mathematical theories like information theory, optimization, functional analysis and signal processing. I enjoy picking up new problems and thrive in unfamiliarity — while I primarily work on machine perception and computer vision, my work extends to problems in robotics, language models and reinforcement learning.

I am currently a member of the perception team at Waymo where I train and deploy real-time AI models for perception. Our team uses millions of miles of driving data to build spatio-temporal representations of the world from multiple sensors. I’ve also previously interned at AWS, Microsoft and the Harvard Center for Brain Science with NTT.

I obtained a PhD in Computer Science from the University of Pennsylvania (2025) and was a member of the GRASP lab. My dissertation proposed theories for multi-task and self-supervised learning and I was advised by Pratik Chaudhari. I also hold a Masters in Robotics from the University of Pennsylvania (2024).

Prior to that, I graduated from IIT Madras at the top of my cohort, with a Dual Degree (Bachelors and Masters) in Computer Science and a minor in Physics (2019). I was advised by Balaraman Ravindran and worked on hierarchical reinforcement learning for my dual degree thesis.