Hyeonwoo Kang
Vision AI Researcher at NC AI
South Korea
Summary
🤩
Rockstar🎓
Top SchoolHyeonwoo Kang is a Vision AI researcher with eight years of hands-on experience building and refining generative and image-to-image translation models in production settings. Having driven vision research at NCSOFT and now at NC AI, he blends deep PyTorch expertise with practical model engineering—contributing notable updates to popular repos like UGATIT and multiple GAN collections. His work spans architecture tweaks, loss-function tuning, and training pipeline improvements, showing strength in turning research ideas into robust code. Based in South Korea and trained at 가톨릭대학교, he is as comfortable debugging low-level network behavior as iterating on dataset preprocessing and evaluation workflows.
8 years of coding experience
8 years of employment as a software developer
가톨릭대학교
Github Skills (12)
computer-vision10
pytorch10
machine-learning10
deep-learning10
cgan10
python10
mnist10
generative-adversarial-network10
dcgan10
resnet9
mask-rcnn9
faster-rcnn9
Programming languages (3)
Github contributions (5)
znxlwm/pytorch-generative-model-collections
Aug 2017 - Jun 2018
Collection of generative models in Pytorch version.
Role in this project:
ML Engineer
ML EngineerContributions:86 commits, 1 PR, 71 pushes in 9 months
Contributions summary:Hyeonwoo primarily contributed to implementing and updating various generative models using PyTorch. Their work involved defining and modifying the architecture of generators and discriminators, with specific implementations like InfoGAN, WGAN, and BEGAN. The user also updated the code to be compatible with Pytorch 0.4 and addressed issues with loss functions within WGAN implementations. The contributions demonstrate a focus on experimenting with and refining different GAN architectures.
pytorchautoencoderdeep-learninggenerative-adversarial-networkgenerative-adversarial-networks
znxlwm/pytorch-MNIST-CelebA-GAN-DCGAN
Jul 2017 - Aug 2017
Pytorch implementation of Generative Adversarial Networks (GAN) and Deep Convolutional Generative Adversarial Networks (DCGAN) for MNIST and CelebA datasets
Role in this project:ML Engineer
ML EngineerContributions:45 commits, 44 pushes, 1 branch in 22 days
Contributions summary:Hyeonwoo primarily contributes to implementing and refining Generative Adversarial Networks (GANs) and Deep Convolutional GANs (DCGANs) using PyTorch. Their commits involve creating and updating the generator and discriminator models, configuring training parameters, and visualizing results for both MNIST and CelebA datasets. They also added a preprocessing step to resize the CelebA images. The contributions include code for model definition, training loops, and result visualization.
pytorchcelebadeep-learningadversarialgenerative-adversarial-network
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.