Sergey Kazakov
Senior Software Engineer at Intel Corporation
Hillsboro, Oregon, United States
Summary
🤩
Rockstar🎓
Top SchoolSergey Kazakov is a Senior Software Engineer with nine years of professional experience, currently contributing to Intel from Hillsboro, Oregon. He specializes in low-level back-end development and performance engineering, with notable contributions to the high-profile oneDNN deep neural network library—optimizing JIT assembler interactions, ISA dispatching for AVX2, and implementing NHWC convolutions for f32/f16. With a Mathematics background from Kazan State University, he brings analytical rigor to CPU-level optimizations and numeric computing. Sergey’s work blends algorithmic understanding with pragmatic microarchitectural tuning, yielding measurable performance gains in production-grade ML kernels. Colleagues rely on him for deep-dive fixes that squeeze extra performance out of existing codepaths while maintaining stability.
9 years of coding experience
Mathematics, Mathematics at Kazan State University
English, Russian
Github Skills (7)
c-language10
cprogramming-language10
performance-optimization10
opencl9
deep-learning9
convolution9
avx9
Programming languages (1)
Github contributions (2)
uxlfoundation/oneDNN
Dec 2016 - Jan 2023
oneAPI Deep Neural Network Library (oneDNN)
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:24 reviews, 241 commits, 3 PRs in 6 years 2 months
Contributions summary:Sergey primarily focused on optimizing code within the oneAPI Deep Neural Network Library (oneDNN) project. Their contributions involved updating the xBYAK JIT assembler library and making specific fixes to vpxor instructions, indicating a focus on low-level code improvements. They also added CPU ISA dispatcher for avx2 kernels and improve performance, highlighting a performance optimization effort. The user also worked on implementing NHWC f32 and f16 convolutions.
bfloat16sse41avx512openmpopencl
skazakov1/DeepBench
May 2017 - May 2017
Contributions:1 push in 1 day
benchmarkingdeep-learningmachine-learningbenchmarkhardware
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.