Developed unified routing system for all recruiting tools at Facebook, enabling over 50 separate applications to function seamlessly as a single-page app. Became a React & ReactRouter expert and coordinated with multiple teams across different offices to make this happen.
Developed C++ client for optimized reading of the internal-to-Facebook fork of the Apache Hadoop ORC file format. Benchmarks showed as much as 50% improvement in CPU time on local file reading speed against existing Java implementation.
Phi Beta Kappa, summa cum laude.
Potentially addictive physics-y infinite runner inspired by the idea of endlessly running down the length of a kaleidoscope. Implemented with ThreeJS.
Worked with Princeton professor Matt Weinberg to prove a theorem showing how an auctioneer running a repeated auction (e.g. a Google ad auction) can exploit bidders running common no-regret learning algorithms to extract full utility from them.
HackMIT winner (Baidu's $10,000 prize for Best Deep Learning Hack). Developed a model for identifying social media comments as containing personal attacks using Baidu's Paddle Deep Learning framework to call attention to online harassment and its biases.
Java, TypeScript, HTML/CSS
React, Haskell, OCaml, Go, Python/NumPy
My team won the Google-sponsored coding/puzzle/trivia/word-association competition twice.