Li Min
Computer Scientist at ISCAS
Beijing, China
Summary
🤩
RockstarLi Min is a computer scientist with 11 years of hands-on experience in high-performance and parallel computing, based in Beijing. He contributes to industrial-grade deep learning infrastructure as a back-end developer and ML engineer on the well-known PaddlePaddle project, focusing on tensor operations, NPU support, and performance-driven refactors like batch normalization optimization. His work bridges low-level operator implementation and practical deployment concerns, showing fluency in both algorithmic detail and production reliability. Comfortable in performance-critical code paths, he emphasizes reusability and cross-platform efficiency when adding features or fixing core-framework issues. Colleagues would note his blend of systems thinking and machine-learning pragmatism—capable of improving throughput without sacrificing maintainability.
11 years of coding experience
Github Skills (18)
gpgpu10
python10
operation10
tensorrt10
gpu-programming10
machine-learning10
n10
deep-learning10
gpu10
cuda10
np10
tensor10
distributed-training9
efficientnet9
cprogramming-language8
Programming languages (3)
Github contributions (5)
PaddlePaddle/Paddle
Jun 2021 - Sep 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:150 reviews, 53 commits, 89 PRs in 1 year 3 months
Contributions summary:Li primarily contributed to the development of PaddlePaddle, a deep learning framework. Their work focused on adding new operations, specifically related to tensor manipulation and NPU support for reciprocal and index selection operations. Furthermore, they were involved in optimizing the batch normalization operation and refactoring existing code to improve reusability and performance. Their contributions included implementing new features, and fixing issues within the core framework.
deep-learningmachine-learningpaddlepaddlescalabilityneural-network
limin2021/TensorRT-LLM
Jun 2025 - Jul 2026
TensorRT-LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and support state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT-LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in performant way.
Contributions:124 pushes, 20 branches in 1 year 1 month
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.