Yujia Zhai

Deep Learning Library Performance Software Engineer at NVIDIA

Milpitas, California, United States
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
🎓
Top School
Yujia Zhai is a deep learning library performance engineer with seven years of experience optimizing compute kernels across multi-core x86 CPUs, GPUs, and heterogeneous platforms. Based in Milpitas, California, Yujia contributes to NVIDIA’s CUTLASS project, bringing hands-on expertise in high-performance primitives like softmax, layer normalization, and fused multi-head attention. Their background blends a PhD in computer science with chemistry degrees, reflecting a cross-disciplinary approach to numerical algorithms and systems-level optimization. Prior internships at ByteDance and Intel produced peer-reviewed IPDPS publications—one earning a Best Paper Award—underscoring an ability to translate research into production-grade GPU acceleration. Notably, Yujia returned to the same NVIDIA team after an internship, signaling strong impact and domain fit in fast kernels and performance engineering.
code7 years of coding experience
job1 year of employment as a software developer
bookMaster of Science - MS, Chemistry, Master of Science - MS, Chemistry at Duke University
bookBachelor of Science - BS, Chemistry, Bachelor of Science - BS, Chemistry at University of Science and Technology of China
bookDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at University of California, Riverside
github-logo-circle

Github Skills (36)

algebra10
matrix-multiplication10
cluster-computing10
blas10
nvidia10
deep-learning10
gpu10
cuda10
cpp10
linear-algebra10
intel9
cpu9
optimization9
rust8
linux8

Programming languages (8)

C++CRustJavaScriptNixRubyPythonCuda

Github contributions (5)

github-logo-circle
yzhaiustc/cutlass

Jul 2022 - Mar 2025

CUDA Templates for Linear Algebra Subroutines
Contributions:51 pushes, 16 branches in 2 years 8 months
cudalinear-algebragpulinearalgebra
Stepwise optimizations of DGEMM on CPU, reaching performance faster than Intel MKL eventually.
Contributions:43 commits, 2 PRs, 9 pushes in 1 year 5 months
cpuoptimizationsintelfasterperformance
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
Yujia Zhai - Deep Learning Library Performance Software Engineer at NVIDIA