Vinnam Kim
Staff Engineer at Qualcomm
Seoul, South Korea
Summary
🤩
Rockstar🎓
Top SchoolVinnam Kim is a Staff Engineer based in Seoul with nine years of experience building autonomous driving and robotic systems, now leading AI efforts at Qualcomm after roles at Intel and Samsung Research. He specializes in C++ and Python development, architecting motion control for SamsungBot and core multi-target tracking and radar–camera fusion for vehicle perception. His work spans embedded real-time control, distributed reinforcement learning in simulation, and production ML tooling—bridging research prototypes to deployable systems. An active open-source contributor, he has enhanced OpenVINO’s neural network compression and expanded Datumaro’s computer vision dataset tooling, improving quantization and visualization for detection and segmentation models. Known for combining systems-level engineering with ML model optimization, he brings hands-on expertise in sensor fusion, model conversion, and dataset visualization that accelerates end-to-end perception stacks.
9 years of coding experience
8 years of employment as a software developer
Bachelor of Engineering - BE Industrial Management Engineering, Bachelor of Engineering - BE Industrial Management Engineering at Korea UniversityMaster of Engineering - MEng Industrial & Systems Engineering, Master of Engineering - MEng Industrial & Systems Engineering at Korea Advanced Institute of Science and Technology
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (19)
data-visualizations10
pytorch10
python10
machine-learning10
data-visualisation10
onnx10
datasets10
deeplearning-ai10
deep-learning10
compression10
quantization10
computer-vision10
data-visualization10
model-conversion10
semantic-segmentation9
Programming languages (6)
Github contributions (5)
open-edge-platform/datumaro
Oct 2022 - Jan 2023
Dataset Management Framework, a Python library and a CLI tool to build, analyze and manage Computer Vision datasets.
Role in this project:
Full-stack Developer
Full-stack DeveloperContributions:25 releases, 746 reviews, 17 commits in 3 months
Contributions summary:Vinnam contributed significantly to the development of the `datumaro` framework. They implemented new visualization features, including label, points, polygon, polyline, and caption visualization. The user also added features for mask, super-resolution, and depth visualization, demonstrating expertise in computer vision dataset management. Furthermore, they improved the performance of mask encoding and addressed documentation aspects, including the rendering of Jupyter Notebook examples.
pythoncomputer-vision-datasetsclipascal-voccoco
openvinotoolkit/nncf
May 2022 - Sep 2022
Neural Network Compression Framework for enhanced OpenVINO™ inference
Role in this project:ML Engineer
ML EngineerContributions:94 reviews, 45 commits, 52 PRs in 4 months
Contributions summary:Vinnam primarily contributed to the development and improvement of the neural network compression framework. Their work focused on enhancing the quantization capabilities, as evidenced by changes to quantization algorithms, metrics, and engine components within the PyTorch framework. They addressed issues related to model conversion, opset versions, and weight sharing in ONNX models, and expanded the framework's support for models, including the addition of examples and tests for a variety of object detection and segmentation models.
bertsemantic-segmentationmixed-precision-trainingtensorflowclassification
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.