Chengyu Dai
Research Engineer at Citadel Securities
Ann Arbor, Michigan, United States
Summary
🤩
Rockstar🎓
Top SchoolChengyu Dai is a Research Engineer with 11 years of experience specializing in distributed training systems for deep learning, currently focused on monetization at Citadel Securities. He was a core developer on BytePS/ByteCCL at ByteDance, delivering up to 4x throughput improvements through GPU pipeline, network, IO, and graph-fusion optimizations for Ads and ASR teams. His background blends a PhD in Physics and CUDA-accelerated molecular dynamics work with practical systems engineering—previous roles include quant trading engineering at Framework Ventures and autonomy/Python/C++ performance work at Uber. Chengyu has a track record of building fault-tolerant, scalable ML infrastructure (from hyperparameter services to communication primitives) and a knack for squeezing real-world performance wins without sacrificing model accuracy.
11 years of coding experience
3 years of employment as a software developer
Doctor of Philosophy - PhD Physics, Doctor of Philosophy - PhD Physics at University of MichiganBachelor's degree Economics, Bachelor's degree Economics at Tsinghua University
Chinese
Stackoverflow
23reputation
421reached
0answers
1question
Github Skills (59)
python10
data-science10
machine-learning10
ml10
deep-learning10
tensorflow10
deep-neural-networks10
neural-network10
simulation9
sdk9
keras9
high-performance9
jupyter-notebook9
gaussian9
quantum-computing8
Programming languages (2)
Github contributions (5)
PhDSelfHelp/mlmono
Sep 2018 - Nov 2018
Mono repo for all machine learning work.
Contributions:24 commits, 23 PRs, 12 pushes in 1 month
pythonmono-repodata-sciencedeep-learningmono
FXuZ/colock-server
Jan 2015 - Apr 2015
Contributions:88 pushes in 3 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.