Since July 2019, I have been a software engineer at Aurora Innovation, where I am solving perception problems for self-driving cars.

Prior to joining Aurora, I received my Ph.D. from the Machine Learning Department at Carnegie Mellon University (CMU), where I was co-advised by J. Andrew Bagnell and Martial Hebert. My research interest is in cost-effective predictions. Specifically, my Ph.D. thesis is on anytime predictors, which can be interrupted at anytime during inference and still produce valid predictions. Furthermore, the more computational cost is consumed before the interruption, the better the predictions are. Hence, anytime predictors can automatically adjust to and utilize any varying test-time budget limits.