Commercial AI Service
VARCO 3D
Led geometry generation and inference, from sparse-voxel models to textured mesh delivery.
Click to see details
Lead 3D AI Research Engineer
Building production-scale 3D generative systems—from native representations and large-model training to CUDA inference and graphics pipelines.
Lead 3D AI Research Engineer
NC AI · VARCO 3D
Hi, I'm Hwan Heo (허환). I work where generative models, systems, and graphics meet, carrying ideas all the way from representations and training code to the meshes people actually use.
At NC AI, I lead research and engineering for the VARCO 3D team, developing production 3D AI systems across large-scale generative model training, inference optimization, and graphics algorithms. I hold both a B.S. and M.S. in Computer Science from Korea University, where I was advised by Professor Hyunwoo J. Kim. Previously, I was a Research Intern at NAVER AI Lab, mentored by Dr. Jin-Hwa Kim.
When I'm away from CUDA kernels and mesh pipelines, I'm usually playing guitar, making music, or following an unexpected STEM rabbit hole. This site is both a record of the systems I've helped ship and a notebook for the ideas I keep exploring.
Led geometry generation and inference, from sparse-voxel models to textured mesh delivery.
Click to see details
Built a 360° capture-to-engine pipeline, optimizing neural rendering for Unity and Unreal.
Click to see details
Developed a two-stage model that carves geometry, then paints high-resolution textures.
Click to see details
Co-developed fast pose refinement for hash-grid NeRF reconstruction, published at ICML 2023.
Click to see details
Built a browser viewer to inspect, edit, train, and export 2D Gaussian Splatting scenes.
Click to see details
Compared VGGSfM and MASt3R for camera recovery in Gaussian Splatting and radiance fields.
Click to see details
Technical Report, Featured on Hugging Face Daily Papers, TLDR AI
The 15th International Conference on ICT Convergence (ICTC), Oct 2024
Oral presentation, domestic conference
The 40th International Conference on Machine Learning (ICML), Jul 2023
preprint, arXiv 2023
Generative AI Seminar (CS.93000B) - KAIST
Machine Learning for Computer Vision - KAIST
K-GEO Festa - Ministry of Land, Infrastructure and Transport in Korea (국토교통부)
National Geographic Information Institute in Korea (국토지리정보원)
NCSOFT
NAVER Cloud
Technical notes on 3D AI, from theory to production:
generative models and large-scale training, neural rendering, graphics algorithms, and web-based 3D tools.
You can also find me on my Korean blog or YouTube channel, where I share guitar performances and arrangements (currently on a break).
AI Research Engineer building production 3D AI systems across generative models, neural rendering, and graphics.
MLV Lab, Korea University, Seoul
Korea University, Seoul
3D Generation, NC AI
Neural Rendering & 3D Generation, NCSOFT Research
Generation Research, NAVER AI Lab