Haihao Shen
Principal AI Engineer at Intel Corporation
Shanghai, Shanghai, China
Summary
🤩
Rockstar🎓
Top SchoolHaihao Shen is a Principal AI Engineer with nine years of experience leading model optimization, compression, and deployment efforts at Intel, where he created and now leads the Intel Neural Compressor and heads the Open Platform for Enterprise AI. He has a strong track record of production-focused ML engineering—contributing to PaddlePaddle core features and INT8 calibration APIs—and has driven benchmarking and optimizations across major frameworks like TensorFlow and PyTorch. Based in Shanghai, he blends deep systems and compiler-aware optimization skills with product leadership, shepherding work from low-level operator improvements to end-to-end GenAI platform integration. Unusually, his background spans formal program analysis and large-scale testing tools, giving him a rigorous approach to reliability and performance in AI systems.
9 years of coding experience
14 years of employment as a software developer
Master, Program analysis and software verification, Master, Program analysis and software verification at Shanghai Jiao Tong University
Github Skills (11)
paddlepaddle10
machine-learning10
deeplearning-ai10
deep-learning10
python10
efficientnet10
neural-network9
distributed-training9
c-language8
cprogramming-language8
cuda7
Programming languages (9)
Github contributions (5)
PaddlePaddle/Paddle
Aug 2018 - Jan 2019
PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）
Role in this project:
ML Engineer
ML EngineerContributions:8 commits, 6 PRs, 161 comments in 5 months
Contributions summary:Haihao contributed to the PaddlePaddle framework, primarily focusing on integrating and optimizing machine learning functionalities. Their work included enabling features like convolution and batch normalization by default, and merging code changes related to LSTM operator enhancements. The user also refined INT8 calibration APIs and integrated performance measurements within the calibration unit tests. This suggests a focus on improving model performance and deployment efficiency.
pytorchpythonparalleldeep-learningpaddlepaddle
hshen14/Paddle
Aug 2018 - Feb 2019
PArallel Distributed Deep LEarning
Contributions:54 pushes, 11 branches in 5 months
caffe2paralleldeep-learningpaddlepaddlemachine-learning
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.