Sheng Jia
Researcher at MiniMax
California, United States
Summary
👤
Senior🎓
Top SchoolSheng Jia is an applied scientist and PhD candidate at the University of Toronto with eight years of experience at the intersection of LLMs, reinforcement learning, and agent design. Currently at Amazon, Sheng focuses on scaling RL for coding models and previously developed SSFT — a self-supervised bipartite-matching post-training method accepted at ICLR 2026 with code and models publicly available. His work spans industry research roles (including contributions to MiniMax-M2.5/M2.7) and long-term research training at the Vector Institute, combining rigorous academic grounding with production-minded ML engineering. Based in Toronto, he brings deep expertise in LLM training dynamics and a knack for turning cutting-edge research into reproducible code and model releases.
8 years of coding experience
2 years of employment as a software developer
Doctor of Philosophy - PhD Computer Science (Machine Learning), Doctor of Philosophy - PhD Computer Science (Machine Learning) at University of Toronto
Japanese, Chinese, EnglishGithub contributions (4)
Sheng-J/DOM-Q-NET
Jan 2019 - Jul 2019
Contributions:21 commits, 19 pushes in 5 months
Sheng-J/scntk
Jun 2021 - Jun 2021
Contributions:1 commit, 1 push, 1 branch in 1 day
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.