Chao C
Software Engineering Manager at AMD
United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolChao C is a Software Engineering Manager at AMD with 11 years of experience building and optimizing ML frameworks, compilers, and GPU backends. He combines hands-on performance engineering—contributing to high-profile open-source projects like TensorFlow and XLA—with leadership of teams shipping ROCm integrations and bfloat16/FP optimizations for CUDA and AMD platforms. His background spans academic research and production: a PhD-trained engineer who built deep-learning toolkits, large-scale ETL pipelines, and HPC workflows at the University of Bristol and Warwick. Known for low-level GPU kernel tuning and cross-platform compiler fixes, he bridges research-grade algorithms and production performance, often surfacing subtle numerical and build issues that improve real-world ML throughput.
11 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy (PhD) Computer Science, Doctor of Philosophy (PhD) Computer Science at University of Warwick
English, Chinese
Github Skills (23)
c-language10
cublas10
gpu-programming10
roc10
tensorflow10
gpu10
performance-optimization10
optimisation10
cuda10
xla10
compiler10
cprogramming-language10
optimization10
testing9
blas9
Programming languages (12)
Github contributions (5)
ROCm/tensorflow-upstream
Jul 2022 - Jan 2023
TensorFlow ROCm port
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:365 reviews, 114 commits, 200 PRs in 6 months
Contributions summary:Chao's commits primarily involve modifying code related to ROCm (AMD's open-source platform for GPU computing), focusing on integrating and optimizing TensorFlow for AMD GPUs. Their work includes fixing build errors, adapting code to support bfloat16 data type, and optimizing performance through modifications in convolution and batch normalization operations. Furthermore, the user refactored code related to launch configurations and low-level instructions to improve efficiency.
rocmtensorflow
openxla/xla
Aug 2022 - Jan 2023
A machine learning compiler for GPUs, CPUs, and ML accelerators
Role in this project:Back-end & Performance Engineer
Back-end & Performance EngineerContributions:69 reviews, 13 commits, 52 PRs in 4 months
Contributions summary:Chao primarily focused on improving the performance and functionality of the XLA compiler, a machine learning compiler. They resolved conflicts related to the movement of `stream_executor`, and contributed to various testing and optimization efforts within the GPU backend. The user's commits included fixing GPU kernel tiling tests, CUDA-specific tests, and addressing warnings and status issues. Furthermore, they added BF16 Matmul support and fixed batch matmul functionality, showing a focus on numerical computation optimization.
compilercommunity-drivenmachine-learningmodular
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.