Minseok Lee
Machine Learning Enginner at MakinaRocks
South Korea
Summary
👤
Senior🎓
Top SchoolMinseok Lee is a machine learning engineer with 8 years of experience specializing in vision and generative models, currently deployed as a Forward Deployed Engineer at MakinaRocks. He built production-grade image tools at Hotpot.ai, including 2D background removal, text-and-image-to-image headshot/art generators, and document QA agents powered by Qwen2.5-VL. Minseok is proficient in customizing VLMs and diffusion models (ControlNet, IPA, cross-attention, SDXL/SD3/FLUX) and in practical model optimization—quantization, pruning, ONNX—targeted for constrained GPUs. His background in industrial and management engineering and early healthcare ML work gives him a strong blend of systems thinking and domain-driven data work. He’s comfortable across the ML stack from dataset curation and SFT/RLHF with PEFT to deployment, with a knack for squeezing high-performance models into real-world resource limits.
8 years of coding experience
5 years of employment as a software developer
Master's degree Industrial and management engineering, Master's degree Industrial and management engineering at Korea UniversityBachelor's degree Business administration, Bachelor's degree Business administration at The Catholic University of Korea
Github Skills (5)
attention-mechanism10
image-segmentation10
tracer10
pytorch8
deep-learning6
Programming languages (1)
Github contributions (5)
Karel911/TRACER
Dec 2021 - Dec 2022
TRACER: Extreme Attention Guided Salient Object Tracing Network (AAAI 2022) implementation in PyTorch
Contributions:34 commits, 3 PRs, 36 pushes in 11 months
pytorchsalient-object-detectionextremeaaai2022attention-mechanism
Karel911/kaggle
Jul 2017 - Jun 2019
Contributions:5 pushes, 1 branch in 1 year 11 months
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.