Sungchul Kim
AI Team Lead at 메딜리티 (PillEye)
South Korea
Summary
🤩
Rockstar🎓
Top SchoolSungchul Kim is an AI Team Lead and research engineer with four years of hands-on experience building and deploying vision AI systems, from real-time inspection (PouchEye/PillEye) to medical imaging and industrial counting. He blends research-grade innovations—such as rapid SAM-based mask generation, efficient SAM fine-tuning, and supervised contrastive learning for small-data regimes—with production MLOps that automate data, training, evaluation, and deployment. At Intel he maintained widely used tooling like OpenVINO Training Extensions and contributed practical compression and domain-adaptation techniques; at startups he translated those advances into product-grade real-time inspection and counting systems. Comfortable across the full model lifecycle, he actively contributes to open-source ecosystems and collaborates across multinational teams to optimize performance. Colleagues rely on him for rapid POCs that surface deployable improvements, and he’s driven by a goal to independently develop, deploy, and operate models across diverse domains.
4 years of coding experience
7 years of employment as a software developer
Master's degree, Biomedical Engineering, Master's degree, Biomedical Engineering at University of UlsanBachelor's degree, Biomedical Engineering, Bachelor's degree, Biomedical Engineering at Yonsei University
Github Skills (57)
pytorch10
text-recognition10
mmsegmentation10
anomaly-detection10
classification10
image-classification10
semantic-segmentation10
openmmlab10
incremental-learning10
machine-learning10
onnx10
openvino10
sift10
super-resolution10
semi-supervised-learning10
Programming languages (2)
Github contributions (5)
sungchul2/training_extensions
Apr 2022 - Oct 2024
Contributions:2 PRs, 602 pushes, 215 branches in 2 years 6 months
optimizationoptimization-toolsmachine-learning
openvinotoolkit/model_preparation_algorithm
Aug 2022 - Dec 2022
Model Preparation Algorithm: a Transfer Learning Framework
Contributions:6 reviews, 87 commits, 5 PRs in 4 months
pytorchsegmentationpreparationdetectiondeep-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.