Jiayi Weng is a research engineer at OpenAI on post-training / RL infrastructure for aligning foundation models. He created the Tianshou PyTorch deep RL library and co-created EnvPool, a high-throughput RL environment executor (NeurIPS 2022). His CV lists contributions spanning ChatGPT's initial release, GPT-4 RL infra, GPT-4o post-training infra lead, and o-series RL infra.123
At a glance
- Current role
- Research engineer
- Company
- OpenAI
- Sources cited
- 3
- Profile reviewed
Accomplishments
OpenAI RL infrastructure
Since July 2022 Weng has built distributed RL training infrastructure at OpenAI. His homepage credits him on ChatGPT (6th author on initial effort notes), GPT-4 RL infra, multimodal RL for GPT-4V, GPT-4o post-training infra lead, and o-series / RFT core RL infra.1
Tianshou and EnvPool
Before OpenAI he created Tianshou (JMLR) and EnvPool (NeurIPS 2022) while at Tsinghua TSAIL, CMU, and Sea AI Lab—tools widely used for reproducible, high-throughput RL research.231
Timeline
Where they work
- OpenAI 93 people in the wiki
Related people
- Jerry TworekAlso at OpenAI
- John SchulmanAlso at OpenAI
- Shantanu JainAlso at OpenAI
- Shengjia ZhaoAlso at OpenAI
- Wojciech ZarembaAlso at OpenAI
- Eiso KantLanguage models · Developer tools
- Omar KhattabDeveloper tools · Language models
- Mika SenghaasReinforcement learning · Language models
- Josh AlbrechtDeveloper tools · Reinforcement learning
- Alessio FanelliLanguage models · Developer tools
Suggest a correction or removal
Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.
References
3 sources · profile reviewed Oct 24, 2026 · data updated Oct 24, 2026