Alvin Huang
Sr. Director - Asus MSI GBT Global Account Lead at AMD
New Taipei, Taiwan
Summary
🎓
Top SchoolAlvin Huang is a results-driven senior sales and business development leader with 20+ years in the high-tech industry and a current role as Sr. Director and Global Account Lead for Asus/MSI/GBT at AMD. He has a strong track record driving revenue, go-to-market strategies, and design-win programs across global OEMs, backed by cross-functional leadership and PMI-PBA project management rigor. His career spans hands-on P&L and territory management roles at ASUS and MSI, giving him deep channel, product, and regional market expertise across APJ, China, and emerging markets. Unusually for a sales executive, Alvin also contributes to open-source GPU performance projects (ROCm/Tensile and MIOpen), bringing technical performance engineering experience that informs his conversations with engineering partners. Based in New Taipei, he blends commercial instincts with technical credibility to align product development, marketing, and global sales execution.
7 years of coding experience
6 years of employment as a software developer
Bachelor Degree, Business administration, Bachelor Degree, Business administration at National Taiwan University of Science and Technology
Github Skills (13)
matrix-multiplication10
opencl10
contraction10
gemfire10
c-language10
cprogramming-language10
gpu10
performance-optimization10
python10
assembly9
machine-learning8
auto-tuning8
hip7
Programming languages (4)
Github contributions (5)
ROCm/Tensile
Jul 2019 - Aug 2021
Stretching GPU performance for GEMMs and tensor contractions.
Role in this project:
Backend Engineer
Backend EngineerContributions:7 reviews, 7 commits, 2 PRs in 2 years 1 month
Contributions summary:Alvin contributed to the core functionality of the Tensile library, focusing on performance and functionality improvements within the GEMM and tensor contraction domain. They implemented new features such as supporting the vpgr limit during training. The user modified several files, including `BenchmarkProblems.py`, `ClientWriter.py`, `TensileCreateLibrary.py`, `Common.py`, and `KernelWriterAssembly.py`, to incorporate these changes and address runtime exceptions. Furthermore, the user updated data types for matrix multiplication operations to improve efficiency.
amdpythontensorhipauto-tuning
ROCm/MIOpen
Jul 2019 - Sep 2019
AMD's Machine Intelligence Library
Role in this project:Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:6 commits in 1 month
Contributions summary:Alvin focused on integrating and optimizing Static Compiled GEMM (SCGEMM) kernels within the MIOPEN library. Their contributions involved the development of SCGEMM API-related code, including the creation of tensor shapes, auxiliary data generation, and the implementation of OpenCL aux kernels. They also addressed bug fixes, performance improvements, and refactoring of the SCGEMM solver class. This work aimed to enhance the performance of machine learning operations by leveraging static compilation techniques.
amdcluster-computingmachine-learningblasmachine-intelligence
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.