Summary
Wangyu Han is a Machine Learning Engineer based in Seattle with 9 years of experience building AI infrastructure and production ML systems at scale. He has driven end-to-end ML and serving platforms across Amazon (AWS Bedrock, Ads Creative X, Tax ML) and NAVER, delivering LLM integrations, routing layers, autoscaling, monitoring, and the first GPT-3 service in Korea. At AWS he led efforts to run Anthropic models on Trainium and to launch Sonnet 3.5, combining low-level performance tuning with high-impact platform design. Comfortable in C++, Python, gRPC and Kubernetes, he’s as focused on operational resilience and cost-efficient inference as on model quality. Colleagues describe him as a pragmatist who hates poor documentation — and therefore builds systems that are observable, well-instrumented, and easy to hand off.
9 years of coding experience