I am a PhD Candidate in the UC Berkeley RISE Lab. My research interests lie at the intersection of systems and machine learning, specifically around how to build systems that make it easier to create fast, safe, affordable, and maintainable applications for serving machine learning at scale. In contrast to much of the work at the intersection of systems and machine learning which focuses on large-scale model training, my work focuses on the model inference part of the machine learning lifecycle.
PhD in Computer Science, 2019 (projected)
University of California, Berkeley
BSc in Computer Science, Physics, 2013
Johns Hopkins University
Clipper is general-purpose, low-latency prediction serving system.
Flor is a system for building, configuring, and tracking machine-learning workflows.
GraphX is Apache Spark’s API for graphs and graph-parallel computation.
Velox is a system for serving machine learning predictions, featuring tight integration with Apache Spark and KeystoneML, model retraining that spans online and batch learning, and automatic support for personalized predictions.