Summary
Kurt Yang is a software engineer with 14 years of experience specializing in large-scale ML infrastructure and high-performance inference systems. Currently at Google on Vertex AI, he previously accelerated TPU LLM training/inference and optimized CPU-based LLM decoding at Oracle’s HeatwaveML, delivering multi-fold speedups. His recent work includes contributions to MaxText on JAX and practical optimizations like speculative decoding that improved token generation throughput. With an academic background spanning a PhD in Computer Science, an MS in Electrical Engineering from Columbia, and a BE from Glasgow, he blends rigorous research with production-grade engineering. Based in the Greater Seattle Area, he focuses on squeezing latency and cost out of ML stacks while enabling scalable model deployment. Colleagues would describe him as a pragmatic optimizer who turns academic insight into measurable system gains.
14 years of coding experience
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at Ventura College
Bachelor's degree, Computer Science, A, Bachelor's degree, Computer Science, A at Portland State University
Computer Information Systems, Computer and Information Sciences, General, A, Computer Information Systems, Computer and Information Sciences, General, A at Portland Community College
Chinese, Chinese Language and Literature, A, Chinese, Chinese Language and Literature, A at BLCU
Chinese