Heyang Qin
Member Of Technical Staff at Anthropic
Reno, Nevada, United States
Summary
🤩
Rockstar🎓
Top SchoolHeyang Qin is a research-focused software engineer with seven years of experience building large-scale model infrastructure and inference systems, currently a Member of Technical Staff at Anthropic. He previously drove model onboarding and EngineV2/3 efforts for Azure OpenAI as a Senior Researcher at Microsoft and contributed to DeepSpeed improvements like ZeRO-3 and ZeRO++ that accelerate distributed training. Holding a PhD in Computer Science from the University of Nevada, Reno, he blends rigorous academic research with production-grade engineering across cloud and ML systems. Based in Reno, he has a track record of moving cutting-edge research into scalable platform features and subtly favors optimizing memory and inference paths—skills that repeatedly enable larger models to run efficiently in real-world deployments.
7 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy - PhD Computer Science and Engineering, Doctor of Philosophy - PhD Computer Science and Engineering at University of Nevada, RenoUniversity of Electronic Science and Technology of China
Stackoverflow
1,574reputation
554kreached
60answers
7questions
python
top-5%
Github Skills (86)
python10
command-line-interface10
deep-learning10
gpu10
compression10
gpu-acceleration10
optimization10
distributed-training10
acceleration10
numpy10
neural-network10
data-parallelism10
pytorch10
mixture-of-experts10
machine-learning10
Programming languages (7)
Github contributions (5)
HeyangQin/RRL
Apr 2019 - Aug 2019
Code for paper Swift Machine Learning Model Serving Scheduling: A Region Based Reinforcement Learning Approach
Contributions:3 commits, 2 pushes, 1 branch in 4 months
servingschedulingapproachreinforcement-learningmachine-learning
HeyangQin/DeepSpeed
Nov 2021 - Nov 2022
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.
Contributions:1 PR, 13 pushes, 10 branches in 1 year
pytorchdeepspeeddeep-learningeffectiveoptimization
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.