Sang Lee
AI Compiler Engineer at Intel Corporation
California, United States
Summary
🤩
Rockstar🎓
Top SchoolSang Lee is an AI compiler engineer with nine years of experience building and optimizing graph compilers and toolchains for production ML workloads, currently working on MLIR/LLVM for Intel AI accelerators and GPUs. He has deep hands-on experience integrating and improving compiler runtimes—contributions include nGraph integration into PaddlePaddle and upstream fixes to Nervana’s ngraph—bridging research-grade compiler work with industrial ML frameworks like ONNX Runtime and oneDNN Graph. Earlier work in binary analysis and translation informs his pragmatic approach to performance and portability across architectures. Holding MS and PhD degrees in Electrical and Computer Engineering from Purdue and a BS from Seoul National University, he combines rigorous academic training with enterprise-scale engineering. Colleagues rely on him for solving tricky cross-platform build and runtime issues that often hide behind flaky builds or mismatched dependencies.
8 years of coding experience
5 years of employment as a software developer
BS, Electrical Engineering, BS, Electrical Engineering at Seoul National UniversityDoctor of Philosophy (PhD), Electrical and Computer Engineering, Doctor of Philosophy (PhD), Electrical and Computer Engineering at Purdue University
Github Skills (18)
c-language10
python10
cmake10
deeplearning-ai10
deep-learning10
build-automation10
paddlepaddle10
cprogramming-language10
distributed-training9
unit-testing9
cicd9
machine-learning9
onnx8
algorithms7
data-structure7
Programming languages (4)
Github contributions (5)
NervanaSystems/ngraph
Nov 2017 - Sep 2020
nGraph - open source C++ library, compiler and runtime for Deep Learning
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1 review, 603 commits, 380 PRs in 2 years 10 months
Contributions summary:Sang contributed to the ngraph project, a C++ library and compiler for deep learning, by addressing compiler errors on Mac and improving the test suite. They added support for constant outputs in the aliased_output test and implemented and tested the TopK operation with Python wrappers. Furthermore, they made several improvements to the build process and added license files.
inference-enginecppc-librarydeep-learningtvm
PaddlePaddle/Paddle
Oct 2018 - Jan 2019
PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice （『飞桨』核心框架，深度学习&机器学习高性能单机、分布式训练和跨平台部署）
Role in this project:MLOps Engineer
MLOps EngineerContributions:5 commits in 2 months
Contributions summary:Sang focused on integrating nGraph, a deep learning compiler, into the PaddlePaddle framework. Their contributions included adding nGraph build instructions, modifying build scripts (CMake and bash), and incorporating nGraph into the inference demo build. They also refactored build settings and fixed build issues on CentOS, demonstrating an understanding of the build process and its dependencies. Further commits involved setting the correct TBB library name in debug builds.
pytorchpythonparalleldeep-learningpaddlepaddle
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.