I am an assistant professor in the EECS department at UC Berkeley and a founding member of the new UC Berkeley RISE Lab. My research interests are at the intersection of machine learning and data systems and explore the challenges of distributed machine learning and inference on large models and datasets, real-time model serving and personalization, and applying machine learning techniques to system tuning and management.
Background: Before joining UC Berkeley as an assistant professor, I was a post-doc in the UC Berkeley AMPLab working on several projects including GraphX (now part of Apache Spark), early versions of MLbase, Velox, and concurrency control for ML. I obtained my PhD from the Machine Learning Department at Carnegie Mellon University where I worked with Carlos Guestrin on Parallel and Distributed Systems for Probabilistic Reasoning.
I am fortunate to advise the following excellent students.
We are looking for postdocs and graduate students to join the new UC Berkeley RISE Lab. As a founding member of the RISE Lab I am working on several projects in large-scale systems for real-time and secure machine learning, time-series analysis, system management, and reinforcement learning. My machine learning research spans both classical statistical machine learning methods as well as new work in deep learning. If you are interested please consider applying to the UC Berkeley CS graduate program in DBMS or AI.
I am currently on the technical advisory board for Deepscale.ai which is developing new computer vision software and systems for autonomous vehicles. Deepscale.ai is currently hiring!
My research as part of the UC Berkeley RISE Lab is funded by a group of 10 core industrial partners.