Tristan Thrush
Doctoral Researcher at Stanford Artificial Intelligence Laboratory (SAIL)
United States
Summary
🤩
Rockstar🎓
Top SchoolTristan Thrush is a doctoral researcher at Stanford with nine years of experience at the intersection of NLP, multimodal computer vision, and reinforcement learning. He has driven research and engineering at top labs and startups including Hugging Face, Meta FAIR, MIT CSAIL, and NASA JPL, contributing to model evaluation, RLHF, continual learning, and large-scale pretraining. A founding technical team member at Contextual AI, he led model evaluation efforts and watched a fast-growing startup scale firsthand. His MIT training (MEng and BS with top GPAs) underpins work ranging from neural RL algorithms to state-of-the-art stereo visual odometry applied to Mars sample recovery. Tristan prioritizes substantive research inquiries—his publications on Google Scholar and Semantic Scholar offer deeper insight into his work—and is selective about outreach, favoring conversations that engage his research interests.
9 years of coding experience
7 years of employment as a software developer
Master of Engineering - MEng, Electrical Engineering and Computer Science, Artificial Intelligence Concentration, 5.0 GPA, Master of Engineering - MEng, Electrical Engineering and Computer Science, Artificial Intelligence Concentration, 5.0 GPA at Massachusetts Institute of TechnologyDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Stanford University
English
Github Skills (59)
face10
pytorch10
robotics10
evaluation10
pandas10
machine-learning10
verification10
reinforcement-learning10
datasets10
lucee10
speech10
deep-learning10
natural-language-processing10
language-modeling10
computer-vision10
Programming languages (4)
Github contributions (5)
huggingface/olm-training
Oct 2022 - Feb 2023
Repo for training MLMs, CLMs, or T5-type models on the OLM pretraining data, but it should work with any hugging face text dataset.
Contributions:1 review, 30 commits, 1 PR in 4 months
TristanThrush/drake
Jun 2017 - Oct 2017
A planning, control, and analysis toolbox for nonlinear dynamical systems. More info at
Contributions:96 pushes, 5 branches in 3 months
structural-engineeringnonlineartoolboxdynamical-systemsanalysis
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.