Hyunho Yeo
I am a Machine Learning (ML) engineer at Moloco, specifically working with the ML-Infra team to build a planet-scale serving system that is highly scalable and cost-effective!
I received my Ph.D. in Electrical Engineering at KAIST, advised by Professor Dongsu Han. During my Ph.D., I have worked on ML systems and published papers at top-tier system conferences such as OSDI, SIGCOMM, MobiCom, and CVPR. Through my research, I have developed expertise in building high-performance ML systems at scale and optimizing the cost of resource-intensive ML models. I am proficient in utilizing machine learning frameworks like PyTorch, TensorFlow, and TensorRT to implement full-fledged end-to-end systems. Before that, I obtained a BS degree in Electrical Engineering from KAIST in South Korea.
selected publications (see all)
-
CVPRAccelIR: Task-aware Image Compression for Accelerating Neural RestorationIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Jun 2023
-
SIGCOMMNeuroScaler: Neural Video Enhancement at ScaleIn Proceedings of the ACM SIGCOMM 2022 conference Aug 2022