Qingyang Li
Head Of RLHF In LLMs And Multi-modality at Kuaishou Technology
Mountain View, California, United States
Summary
🤩
Rockstar🎓
Top SchoolQingyang Li is a research-driven engineering leader with a PhD in Computer Science and nine years of experience building ML systems across autonomous driving and multimodal LLMs. Currently Head of RLHF in LLMs and Multi-modality at Kuaishou, he leads RLHF, RM, DPO and PPO efforts to align and scale generative models and supports captioning for Kling. Previously he led ML research and planning teams at DiDi, bringing imitation learning and production-grade planning modules to autonomous vehicles. Qingyang blends deep academic roots in ML, RL and computer vision with hands-on production experience at Amazon, Intel and large-scale platforms, and his GitHub motto "fast is slow" hints at a methodical focus on robust, well-tested systems. Based in Mountain View, he is known for turning cutting-edge research into deployable systems that balance innovation with safety and alignment.
9 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy (PhD) Computer Science, Doctor of Philosophy (PhD) Computer Science at Arizona State UniversityBachelor's degree Computer Software Engineering, Bachelor's degree Computer Software Engineering at Beihang University
Github Skills (32)
debugging10
debug10
jvm10
monitor10
cpu10
java10
agent10
alibaba10
java-agent10
trace10
diagnostic-tool9
rundeck9
sampling9
glibc9
perf9
Programming languages (2)
Github contributions (5)
wenbeicai/pages
Oct 2016 - Oct 2016
Contributions:15 pushes, 2 branches in 1 day
wenbeicai/HTTPAttack
Mar 2017 - Apr 2017
Contributions:8 pushes, 2 branches in 1 month
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.