Xu Xianming

Senior Software Engineer at Baidu, Inc.

China
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
Xu Xianming is a Senior Software Engineer based in China with eight years of experience building high-performance back-end systems and machine learning infrastructure. At Baidu and as a contributor to the prominent PaddlePaddle project, he has focused on GPU compatibility and performance, notably resolving ROCm/HIP issues and optimizing CUDA kernels and memory paths for AMD GPUs. His work improved core operations like memcpy and extended CPU support for seed and dropout operators, contributing measurable performance gains to an industrial-grade deep learning framework. Comfortable at the intersection of systems engineering and ML, he blends low-level optimization with practical model training needs. Colleagues would describe him as a pragmatic problem-solver who seeks measurable impact in large open-source and enterprise codebases.
code8 years of coding experience
github-logo-circle

Github Skills (13)

cuda10
paddlepaddle10
hip10
machine-learning10
deeplearning-ai10
deep-learning10
amdgpu10
python10
efficientnet10
cprogramming-language9
neural-network9
distributed-training9
c-language9

Programming languages (4)

ShellC++GoPython

Github contributions (5)

github-logo-circle
PaddlePaddle/Paddle

Feb 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:
userBack-end Developer & ML Engineer
Contributions:73 reviews, 20 commits, 35 PRs in 1 year 2 months
Contributions summary:Xu primarily contributed to the PaddlePaddle deep learning framework by addressing bugs related to ROCm/HIP and AMD GPU compatibility, significantly improving performance. Their work involved modifications to CUDA kernels, and implementation of memory optimization techniques for AMD GPUs, with a stated performance gain. Furthermore, the user improved functionality by resolving issues with the memcpy operation and integrating support for the seed and dropout operators on the CPU for enhanced model training.
pytorchpythonparalleldeep-learningpaddlepaddle
freeliuzc/PaddleCloud

May 2020 - Jul 2020

PaddlePaddle Docker images and K8s operators for PaddleOCR/Detection developers to use on public/private cloud.
Contributions:239 commits in 2 months
nvidia-dockerpaddlepaddlepaddleocrdockerk8s
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.
Request Free Trial
Xu Xianming - Senior Software Engineer at Baidu, Inc.