Eric Chung

Eric Chung is vice president of AI Computing at NVIDIA, where he leads efforts to scale AI through breakthroughs in model efficiency, low-precision numerics, software resiliency, and agent technologies. Most recently, his teams created and deployed Puzzle, a distillation-driven neural architecture search framework used to optimize inference in NVIDIA state-of-the-art Nemotron reasoning models. Previously, Eric led AI supercomputing and hardware-software co-design efforts at Microsoft, where his team co-founded the OCP MX Alliance and introduced the MXFP formats now deployed ubiquitously across AI hardware, including NVIDIA Blackwell. He holds a PhD in Electrical and Computer Engineering from Carnegie Mellon University.
Avatar photo

Posts by Eric Chung

Data Center / Cloud

Introducing NVFP4 for Efficient and Accurate Low-Precision Inference

To get the most out of AI, optimizations are critical. When developers think about optimizing AI models for inference, model compression techniques—such as... 11 MIN READ