|Sept 2018||Delphi will appear as a NIPS 2018 spotlight: Infer loop invariants by reinforcement learning!|
|May 2018||We will be presenting an overview of the Difflog program reasoning framework at MLP 2018!|
|Feb 2018||Bingo will appear at PLDI 2018: Human-in-the-loop + Bayesian reasoning ⇒ Dramatically more accurate bug finding!|
I am a postdoc working with Mayur Naik at the University of Pennsylvania.
My research is at the intersection of programming languages, software engineering and automated reasoning. I draw on techniques from machine learning and formal methods to solve problems in program synthesis, verification, and static analysis. My goal is to build theoretically well-understood, rigorously evaluated, and practically useful tools to help programmers create better software with less effort.
I was previously a Ph.D. student in the same department, advised by Rajeev Alur, where I contributed to formalizing the SyGuS synthesis framework, and designed domain-specific languages and programming abstractions for stream processing.