Chang Cui
Software Engineer at NVIDIA
Beijing, China
Summary
👤
Senior🎓
Top SchoolChang Cui is a Member of Technical Staff with nine years of experience specializing in heterogeneous computing, high-performance inference, and compiler-level optimizations. He built production TensorFlow→TensorRT and PyTorch→TensorRT converters, hand-wrote CUDA operators, and architected a TensorRT-based GPU inference backend with embedding cache and real-time model switching that runs on 15,000+ GPUs for large-scale advertising. His work at Tencent delivered constrained decoding and SmoothQuant solutions for low-latency, controllable ad inference, and he continues to focus on OS, compiler, and HPC problems. A Peking University MS in High Performance Computing and a long-standing Linux/Vim/open-source habit inform his pragmatic, systems-first approach. Notably, he blends research-grade toolchain development with production deployment experience—bridging graph optimization research and massive, low-latency serving systems.
10 years of coding experience
5 years of employment as a software developer
Master of Science - MS High Performance Computing, Master of Science - MS High Performance Computing at Peking University
Github Skills (72)
vpn9
network-analysis9
censorship-circumvention9
proxy9
relay9
http-proxy9
juniper9
network-programming9
quic9
linux8
golang8
static-server8
encryption8
lua8
windows8
Programming languages (8)
Github contributions (5)
hulin12138/zeroOutSocial
Jan 2019 - Jan 2019
Contributions:54 commits, 23 pushes in 2 days
changcui/ZeroutRecorder
May 2017 - Aug 2018
Contributions:3 PRs, 11 pushes, 1 branch in 1 year 2 months
reactrecordingrecording-app
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.