Member of Technical Staff, Mechanize
Evals and RL tasks.
CS at Caltech, interested in Machine Learning and Software Engineering.
Evals and RL tasks.
Related to debugging simulations.
LLM judge evals, automated evalset generation, prompt optimization, and miscellaneous agent engine work.
Guided student projects building a physics engine with collision and SDL graphics, and a video game.
Cross-domain video action recognition with generative diffusion models' attention maps.
CUDA autoregressive inference engine for Qwen3.5 with flash attention and Gated DeltaNet linear-attention layers. CS 179 final project.
Reinforcement learning bot for Settlers of Catan trained with DAgger, PPO, and AlphaZero (WIP). ~70% win rate vs Catanatron baseline bots.
LLM with retrieval-augmented generation (scraped course catalog and 2 years of student feedback) that answers Caltech course questions.
B.S. Computer Science