Summary
Jue Wang is a Principal Software Engineer with nine years in the industry and a deep track record building and optimizing large-scale cloud and ML infrastructure at AWS and Google. He has led major efforts including EC2 P6 instances, virtualization of the GraceBlackwell AI supercomputer, scaling NCCL/EFA to 16,384 H100 GPUs, and shipping Cloud TPU Pods to GA. Author of 15 patents and two defensive publications across cloud computing, virtualization, memory management, distributed systems, and ML, he combines low-level kernel and hypervisor expertise with system-level performance engineering. His work spans product launches, kernel upgrades across millions of Nitro-enabled servers, and power and reliability optimizations for AMD and Intel instances. Based in Greater Seattle, he’s equally comfortable driving hands-on performance measurements and coordinating cross-team integrations that unlock multi-percent performance and megawatt-level power savings. A lesser-known strength is his history of turning research ideas into production features—from memory tiering and virtual IOMMU to the first AMD GPU support on a major cloud.
9 years of coding experience
14 years of employment as a software developer
University of Washington
Tsinghua University