Surya Vengadesan

I am a problem solver broadly interested in machine learning and systems.

I am interested in problems in AI Safety and Reinforcement Learning.

I developed hardware performance tools in Annapurna Labs at AWS.

I developed LLM inference systems in Catalog Systems at Amazon.

I am pursuing an M.S. in CS from UW Seattle.

I earned a B.S. in EECS from UC Berkeley.

Email  /  Google Scholar  /  Github  /  Substack  /  Alignment Forum

We explore the scheduling trade-offs of job completion time and cost.

I attempt to break down an Imitation Learning paper.

I define Behavioral Cloning and train OpenAI's cartpole agent.

I define Markov Decision Processes and implement two algorithms.

I share my experience exploring compilers through CS 164.

I taught math circle courses for high school students on proof writing and game theory.

I taught a student-run course on the societal, economical, and financial impact of AI.