Namgoo Lee
Principal Software Engineer at Stealth Startup
South Korea
Summary
👤
Senior🎓
Top SchoolNamgoo Lee is a Principal Software Engineer with 10 years of experience building high-performance computer vision and industrial AI systems, currently developing algorithms for sheet metal manufacturing and local LLM evaluation on DGX Spark. He progressed from product engineering at a startup to leading CNN-based deep learning maintenance at Cognex, where he earned multiple awards for impactful contributions. An OpenCV contributor, Namgoo has fixed CUDA and Farneback optical flow bugs and optimized GPU code, demonstrating deep expertise in C++, CUDA, and multi-threaded performance tuning. He has a strong track record of practical engineering wins—fp16 quantization, dramatic inference and training speedups, and a near 1000x disk-space reduction concept—combined with automation skills in FreeCAD scripting and CI for cross-platform deployments. Based in South Korea with an MS in Computer Science from Seoul National University, he blends research-level vision work with hands-on productionization in industrial settings.
9 years of coding experience
7 years of employment as a software developer
Master's degree, Computer Science, Master's degree, Computer Science at Seoul National University
Korean, English
Github Skills (13)
cuda10
gpu-programming10
algorithms10
computer-vision10
opencv10
c-language10
cprogramming-language10
optical-flow10
image-processing10
test-automation9
cplus9
cpp9
deep-learning6
Programming languages (4)
Github contributions (5)
opencv/opencv_contrib
Jan 2019 - Jun 2022
Repository for OpenCV's extra modules
Role in this project:
Backend Developer & Test Automation Engineer
Backend Developer & Test Automation EngineerContributions:3 reviews, 16 commits, 3 PRs in 3 years 5 months
Contributions summary:Namgoo's contributions primarily involve fixing bugs and improving the Farneback Optical Flow Algorithm within the OpenCV library. They addressed issues related to the algorithm's handling of the `OPTFLOW_USE_INITIAL_FLOW` flag, ensuring correct behavior and improving test limits. Furthermore, they implemented and tested optimizations for the CUDA implementation, including support for CUDA versions greater than or equal to 9, specifically concerning block scan functions and shared memory usage. These modifications involved code changes in C++ and CUDA, showcasing a strong understanding of GPU programming and image processing algorithms.
pythoncmakecppcomputer-vision
opencv/opencv
Jan 2018 - May 2022
Open Source Computer Vision Library
Role in this project:Back-end Developer & Computer Vision Engineer
Back-end Developer & Computer Vision EngineerContributions:16 reviews, 32 commits, 29 PRs in 4 years 4 months
Contributions summary:Namgoo primarily focused on bug fixes and performance improvements within the OpenCV library. They addressed issues in the CUDA modules, particularly related to color conversion and stream management. Several commits involved fixing defects in core image processing algorithms, such as the Farneback optical flow, and ensuring their compatibility with CUDA. The user also contributed to documentation updates, including examples and explanations for various functionalities.
pythonvisiondeep-learningc-plus-pluscomputer-vision
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.