Baochun Li
Professor at University of Toronto
Old Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolBaochun Li is a Professor of Computer Science at the University of Toronto with over two decades in academia and nine years of documented industry-grade software experience, blending deep research credentials (M.S./Ph.D. from UIUC) with hands-on engineering. He contributes to high-performance machine learning infrastructure, notably implementing NPU operators and rigorous unit tests for the widely used PaddlePaddle framework, signaling practical expertise in accelerating deep learning on specialized hardware. Based in Old Toronto, he bridges theoretical foundations from Tsinghua and UIUC with production-focused backend and ML engineering, comfortable modifying C++ cores and Python test suites alike. His profile reflects a rare mix of long-term academic leadership and modern open-source systems engineering, making him adept at translating research advances into robust, deployable ML primitives.
9 years of coding experience
Jingshan SchoolM.S., Ph.D., Computer Science, M.S., Ph.D., Computer Science at University of Illinois Urbana-ChampaignB. Engr., Computer Science, B. Engr., Computer Science at Tsinghua University
Github Skills (9)
np10
paddlepaddle10
python10
n10
cprogramming-language9
c-language9
deep-learning9
distributed-training8
machine-learning8
Programming languages (4)
Github contributions (5)
PaddlePaddle/Paddle
Aug 2021 - Apr 2022
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:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:92 reviews, 69 commits, 119 PRs in 8 months
Contributions summary:Baochun contributed significantly to the implementation of NPU (Neural Processing Unit) operators within the PaddlePaddle framework, focusing on comparison and elementwise operations. Their work included the addition of `not_equal` NPU operators, as well as the `elementwise max grad` and `strided_slice_grad` ops for the NPU. Furthermore, the user added unit tests and related converter test cases, demonstrating a focus on ensuring the functionality and compatibility of the implemented operators. This involved modifications to both C++ code and Python test scripts.
deep-learningmachine-learningpaddlepaddlescalabilityneural-network
baoachun/A-list-of-popular-compact-models
Dec 2018 - Jan 2021
a list of awesome papers on deep model ompression and acceleration
Contributions:46 commits, 41 pushes in 2 years 2 months
deep-learningpytorchacceleration
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.