Jinfeng Li
Senior Software Engineer, Distributed Machine Learning at NVIDIA
Mountain View, California, United States
Summary
👤
Senior🎓
Top SchoolJinfeng Li is a Senior Software Engineer specializing in distributed machine learning, with 12 years of experience combining research and production engineering across academia and industry. He holds a PhD in Computer Science from The Chinese University of Hong Kong and has translated that research depth into practical systems work at NVIDIA and Megagon Labs. Skilled in NLP, large-scale similarity search, distributed computing, and C++/Java on Linux, he has hands-on expertise building multi-node, multi-GPU ML components—contributing to cuML’s MNMG logistic regression to match scikit-learn behavior for sparse and multi-class scenarios. Based in Mountain View, he excels at bridging algorithmic innovation and robust distributed implementations, and his background in building graph ingestion and indexing systems at Alibaba shows a knack for optimizing data pipelines at scale.
12 years of coding experience
7 years of employment as a software developer
Bachelor of Engineering (B.Eng.) Faculty of Computer Science, Bachelor of Engineering (B.Eng.) Faculty of Computer Science at Sun Yat-sen UniversityThe Chinese University of Hong Kong (CUHK)
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (12)
cuda10
logistic-regression10
machine-learning10
machine-learning-algorithms10
cuml10
gpu10
python9
cprogramming-language9
c-language9
sparse-matrix8
dask8
pytest7
Programming languages (7)
Github contributions (5)
rapidsai/cuml
May 2023 - Dec 2024
cuML - RAPIDS Machine Learning Library
Role in this project:
ML Engineer
ML EngineerContributions:54 reviews, 20 PRs, 78 comments in 1 year 6 months
Contributions summary:Jinfeng primarily contributed to the development and enhancement of multi-node-multi-GPU (MNMG) Logistic Regression within the cuML library. Their work involved implementing the predict API and supporting regularization and standardization for sparse vectors. They also added multi-class support and addressed issues related to all-zero feature vectors, ensuring the MNMG Logistic Regression aligns with scikit-learn behavior.
cudacumlnvidiadata-sciencegpu
lijinf2/gqr
Jan 2017 - Jan 2019
Contributions:258 commits, 26 PRs, 149 pushes in 2 years
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.