Zihao Zhang
Intel MKL Software Engineer at Oregon State University
Portland, Oregon, United States
Summary
👤
Senior🎓
Top SchoolZihao Zhang is a software engineer at Intel with nine years of experience and over six years focused on designing high-performance numerical libraries for MKL. Based in Portland, he blends applied mathematics (MS) and a strong quantitative background (BS in Economics/Mathematics) to optimize compute-heavy workflows and compiler-level numeric transformations. He contributes to open-source ML tooling—enhancing post-training quantization and ONNX op support for the kendryte/nncase deep learning compiler—bringing practical accelerator experience to software stacks. Alongside industry work, he teaches and conducts research at Oregon State University, demonstrating a commitment to mentoring and translating theory into production-grade code.
9 years of coding experience
Canton High SchoolMaster of Science - MS, Applied Mathematics, 3.5, Master of Science - MS, Applied Mathematics, 3.5 at Oregon State UniversityBachelor of Science (BS), Economics/Mathematics, 3.75, Bachelor of Science (BS), Economics/Mathematics, 3.75 at Michigan State University
English, Chinese, Chinese
Github Skills (7)
c-language10
deep-learning10
cprogramming-language10
k210
neural-network9
onnx8
python7
Programming languages (7)
Github contributions (5)
kendryte/nncase
Aug 2021 - Jan 2023
Open deep learning compiler stack for Kendryte AI accelerators ✨
Role in this project:
ML Engineer
ML EngineerContributions:28 reviews, 35 commits, 77 PRs in 1 year 5 months
Contributions summary:Zihao contributed to the development of the deep learning compiler stack, primarily focusing on enhancing post-training quantization features for specific Kendryte AI accelerators. They added new weight quantization types and implemented corresponding support within the compiler. The user also worked on supporting new onnx ops, and improving the code quality and performance of the project.
deep-learningk210kendrytemachine-learningcompiler
aaltonenzhang/caffe
Jul 2021 - Aug 2021
Caffe: a fast open framework for deep learning.
Contributions:189 pushes, 2 branches in 28 days
pytorchmxnetcaffe2deep-learningmachine-learning
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.