Hyunjae Woo

Hyunjae Woo is a senior systems software engineer on NVIDIA's Dynamo team, working on distributed LLM inference systems. His current work focuses on fast model weight transfer across the inference stack, using fast fabrics like InfiniBand and P2P RDMA to push the hardware to its limits. Hyunjae holds a master's and bachelor's from the University of Michigan.
Avatar photo

Posts by Hyunjae Woo

Agentic AI / Generative AI

ModelExpress: Distributing Model Artifacts at the Speed of Light

Every byte moved has a cost. As model checkpoints grow to hundreds of gigabytes or even a terabyte, that cost adds up quickly. To make things even worse, moving... 12 MIN READ