Mickel Liu
Visiting Researcher at Meta
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolMickel Liu is a PhD computer scientist and visiting researcher at Meta FAIR specializing in training and aligning large language models, with seven years of experience spanning RLHF, supervised fine-tuning, and multi-agent research. He bridges academic rigor and industry impact—currently a UW PhD student and research assistant while contributing to open-source releases like Baichuan-7B/13B during an intern role at Baichuan Intelligent Technology. His background in chemical engineering and early work in process control and data analytics gives him a systems-minded approach to model safety and deployment. Mickel has collaborated with AI2 and BIGAI on multi-agent and reinforcement learning projects, and is known among peers as an “academic nomad” based in Seattle who moves fluidly between research labs and product-focused ML teams.
7 years of coding experience
1 year of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at University of WashingtonSecondary Education, International Baccalaureate, Secondary Education, International Baccalaureate at Canadian International School of BeijingMaster of Science - MS, Artificial Intelligence, Master of Science - MS, Artificial Intelligence at Peking UniversityBachelor of Applied Science - BASc, Chemical Engineering, Bachelor of Applied Science - BASc, Chemical Engineering at University of Toronto
English, Chinese
Github Skills (91)
transformers10
safety-critical10
deep-learning10
image-generation10
speech10
optimization10
ai10
hyperparameter-tuning10
gpt10
raylib10
llama10
reinforcement-learning10
amd10
data-versioning10
cuda10
Programming languages (6)
Github contributions (5)
PKU-Alignment/beavertails
Jun 2023 - Jun 2024
BeaverTails is a collection of datasets designed to facilitate research on safety alignment in large language models (LLMs).
Contributions:2 PRs, 1 push, 1 branch in 1 year
large-language-modelsai-safetyhuman-feedbackhuman-feedback-datalanguage-model
mickelliu/OpenRLHF
Apr 2024 - Jun 2024
An Easy-to-use, Scalable and High-performance RLHF Framework (Support 70B+ full tuning & LoRA & Mixtral & KTO)
Contributions:19 pushes, 9 branches in 2 months
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.