Yawen Yang
Graphics Software Engineer at Intel Corporation
Shanghai, Shanghai, China
Summary
👤
Senior🎓
Top SchoolYawen Yang is a Graphics Software Engineer with nine years of experience building video decode SDKs, drivers, and deep learning tooling at Intel after an early engineering role at Fujitsu. Based in Shanghai, she blends low-level graphics and driver development with compiler- and backend-focused ML engineering, contributing notable optimizations to the widely used Apache TVM project (multi-blocking layouts, bfloat16 support, and DNNL BYOC enhancements). Her work demonstrates a pragmatic balance of performance engineering and algorithmic insight, enabling efficient data layouts and in-place operations for production workloads. Comfortable across hardware-near code and high-level ML compiler stacks, she brings cross-domain fluency that helps move models from prototype to optimized deployment.
9 years of coding experience
3 years of employment as a software developer
Shanghai Jiao Tong University
Github Skills (12)
bfd10
compiler10
relayer10
machine-learning10
deeplearning-ai10
c-language10
compiler-compiler10
deep-learning10
cprogramming-language10
optimization10
relay10
python8
Programming languages (4)
Github contributions (5)
apache/tvm
Feb 2022 - Oct 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Role in this project:
Back-end Developer
Back-end DeveloperContributions:15 reviews, 8 commits, 9 PRs in 8 months
Contributions summary:Yawen made significant contributions to the TVM compiler stack, focusing on the backend and optimization aspects. Their work includes adding support for multi-blocking layout transformations, enhancing bfloat16 support in the compiler and relay modules, and integrating in-place post-op sum functionality within the DNNL BYOC backend. They also contributed to improving the handling of data layouts and performing simplification of expressions within the compilation pipeline.
metalvulkancompilertensoropencl
yangulei/tvm
Nov 2021 - Oct 2022
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Contributions:19 reviews, 8 PRs, 120 pushes in 11 months
cpugpu-programminggpu-accelerationtvmdeep-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.