Summary
Govind Pimpale is an AI researcher-engineer based in San Jose with nine years of experience building scalable ML systems, specializing in evaluation and governance of large language models. Currently at hud (YC W25), he builds infrastructure for LLM evaluations and designs coding assessments, drawing on prior research into agent scaling laws and practical improvements to LLM agent tooling. His work spans production systems and research: from CUDA/PyTorch gaze-prediction models with sub-degree error to creating offline RL datasets and developer-facing UI features that wire real-time experiment feedback. Comfortable across stacks (C++, Rust, React, Java, Postgres) and motivated by extracting more signal from evaluations, he blends careful empirical methodology with hands-on engineering to move models from investigation to reliable benchmarking.
9 years of coding experience
5 years of employment as a software developer
Santa Teresa High School
Bachelor's degree, Computer Science and Engineering, Bachelor's degree, Computer Science and Engineering at University of California, Los Angeles