Chuanqi Wang
AI Frameworks Engineer at Intel Corporation
Hefei, Anhui, China
Summary
🤩
Rockstar🎓
Top SchoolChuanqi Wang is an AI Frameworks Engineer at Intel with a decade of experience building and productionizing ML infrastructure and model tooling. He specializes in CI/CD and environment engineering for major deep learning projects—contributing to PyTorch CI to enable XPU (Intel GPU) builds, dynamic CPU accuracy tests, and Triton XPU wheel support—while also improving model compatibility and quantization workflows in Intel’s AI reference models and Neural Compressor. With an MEng in Control Engineering from USTC, he blends control-theory rigor with practical ML engineering to solve stability and edge-case issues in Bayesian tuning and low-bit quantization. Based in Hefei, he is a hands-on engineer who moves models from research to robust, cross-framework deployment.
10 years of coding experience
Master of Engineering - MEng, Control Engineering, Master of Engineering - MEng, Control Engineering at University of Science and Technology of China
Github Skills (29)
python10
machine-learning10
bash10
large-language-models10
cicd10
bayesian10
deep-learning10
tensorflow10
gpu10
optimisation10
build-automation10
neural-network10
quantization10
optimization10
pytorch9
Programming languages (8)
Github contributions (5)
intel/ai-reference-models
Aug 2019 - Nov 2020
Intel® AI Reference Models: contains Intel optimizations for running deep learning workloads on Intel® Xeon® Scalable processors and Intel® Data Center GPUs
Role in this project:
ML Engineer
ML EngineerContributions:1 release, 54 commits, 1 PR in 1 year 2 months
Contributions summary:Chuanqi's commits primarily involve modifications to deep learning models within the repository, focusing on TensorFlow. They addressed issues related to TensorFlow 2.0 API compatibility and debug code, which suggest a focus on maintaining and improving the model's functionality. Contributions also include integrating new model architectures (e.g., SSD-MobileNet) and incorporating the latest updates, indicating a role in model development, integration, and maintenance.
optimizationsprocessorstensorflowzoomodel-zoo
intel/neural-compressor
Jun 2020 - Mar 2021
SOTA low-bit LLM quantization (INT8/FP8/INT4/FP4/NF4) & sparsity; leading model compression techniques on TensorFlow, PyTorch, and ONNX Runtime
Role in this project:ML Engineer
ML EngineerContributions:58 commits in 9 months
Contributions summary:Chuanqi contributed to the Intel Neural Compressor project by enhancing the save/resume functionality for MSE and Bayesian tuning strategies. They addressed bayes restore and corner case issues within the Bayesian strategy, refining its stability. Additionally, the user integrated fixes related to MXNet adaptor and MSE strategies, ensuring compatibility and resolving potential bugs within the quantization process for different machine learning frameworks.
knowledge-distillationauto-tuningcompressorsparsityintel
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.