Sungyeob Yoo
Researcher at KAIST CastLab
Daejeon, South Korea
Summary
👤
Senior🎓
Top SchoolSungyeob Yoo is a Ph.D. candidate in Electrical and Electronics Engineering at KAIST specializing in hardware–software co-design for efficient generative AI acceleration, with expected graduation in 2026. He designs end-to-end AI accelerator solutions—from model-aware numerical methods and architecture-level optimizations to RTL, FPGA prototyping, and silicon validation—targeting diffusion models and transformer workloads. His work emphasizes low-precision quantization, memory-efficient dataflows, and energy- and performance-aware designs validated on real chips and FPGA platforms. At Rebellions he led FPGA-driven evaluations of the ION chip and built an HFT-integrated FPGA system, authoring work on DVFS and task allocation to boost trading performance. With eight years of experience and a track record of shipping custom accelerators, he blends rigorous academic research with practical silicon-prototyping expertise. Based in Daejeon, he brings uncommon depth in bridging numerical model choices to tangible hardware constraints.
8 years of coding experience
Bachelor, Electrical and Computer Engineering, Bachelor, Electrical and Computer Engineering at Ajou UniversityPh.D. Candidate, Electrical and Electronics Engineering, Ph.D. Candidate, Electrical and Electronics Engineering at Korea Advanced Institute of Science and Technology
Github Skills (37)
physics-engine10
pytorch10
pytorch3d10
computational-physics10
photogrammetry10
neural-rendering10
fluid-simulation10
computer-graphics10
voxel10
computational-fluid-dynamics10
cpp10
3d-reconstruction10
3d-graphics9
machine-learning9
3d9
Programming languages (5)
Github contributions (5)
DveloperY0115/DveloperY0115.github.io
Jul 2023 - May 2026
Contributions:117 pushes, 1 branch in 2 years 10 months
DveloperY0115/torch-NeRF
Jun 2021 - Aug 2022
Pytorch implementation of NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis (Mildenhall et al., ECCV 2020 Oral, Best Paper Honorable Mention).
Contributions:3 releases, 418 commits, 11 PRs in 1 year 1 month
pytorchneural-renderingimage-based-renderingnerf
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.