Zhang Xianyi
Co Founder at PerfXLab Technologies Co., Ltd.
Haidian District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolZhang Xianyi is a co-founder and performance-focused systems engineer with 15 years of experience building high-performance numerical and parallel software from research labs to startup productization. With a PhD in Parallel Algorithms and a history of postdoctoral and research fellow roles at MIT and UT Austin, he specializes in CPU architecture-aware optimization—evidenced by contributions that added x86 auto-detection and ARM/NEON support to BLIS and build-kernel work across OpenBLAS. At PerfXLab he bridges deep technical expertise and entrepreneurial execution, turning low-level performance engineering into deployable solutions. He also contributed CPU-native ops and CUDA-conditional builds to the Minerva deep-learning stack, showing a practical blend of ML and systems skills that few performance engineers combine. Based in Haidian, Beijing, he brings academic rigor to production-grade, cross-architecture optimization and CI-enabled open-source workflows.
15 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Parallel Algorithm and Parallel Software, Doctor of Philosophy (Ph.D.), Parallel Algorithm and Parallel Software at University of Chinese Academy of SciencesMaster, Embedded System, Master, Embedded System at Beijing Institute of Technology
Github Skills (23)
assembly10
c-language10
activation-function10
cluster-computing10
machine-learning10
build-system10
blas10
c1110
c1710
deep-learning10
parallel-computing10
cuda10
web-framework10
cprogramming-language10
scientific-computing10
Programming languages (10)
Github contributions (5)
OpenMathLib/OpenBLAS
Jan 2011 - Jan 2023
OpenBLAS is an optimized BLAS library based on GotoBLAS2 1.13 BSD version.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:7 releases, 935 commits, 167 PRs in 12 years 2 months
Contributions summary:Zhang's commits focus on modifying the codebase to align with the evolving requirements of the OpenBLAS project. The changes revolve around updating the build configuration of the source code for multiple architectures. The user's work appears to be centered around building and configuring kernels. The changes include modifying the flags that control the build behavior of the code for a wide variety of architectures including ARM, zarch and x86-64.
lapackopenblaslapackelinear-algebrabsd
flame/blis
Apr 2015 - Oct 2015
BLAS-like Library Instantiation Software Framework
Role in this project:Back-end Developer / Performance Engineer
Back-end Developer / Performance EngineerContributions:8 commits, 3 PRs, 5 comments in 6 months
Contributions summary:Zhang primarily focused on enhancing the BLIS library's CPU architecture detection capabilities to optimize performance. They implemented an auto-detecting mechanism for x86 CPUs and extended support for ARM architectures. The user's contributions also included detecting specific CPU cores and features like NEON for ARM processors. Further contributions involved integrating Travis CI for automated testing and merging updates from the upstream master branch.
software-frameworklinear-algebra-librarylinear-algebramatrix-multiplicationblis
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.