Summary
Zhi-gang Liu is an AI/ML accelerator architect with seven years focused on designing high-performance hardware and systems for machine learning, currently shaping next-generation accelerators at Tenstorrent. His career traverses leading-edge silicon and architecture teams at Samsung Semiconductor, Arm, and NVIDIA, with early experience in multicore and embedded systems from Tilera, EZchip, and EMC. He combines deep physics and electrical engineering training from Tsinghua, Brandeis, and UMass Amherst with pragmatic product-driven engineering, bridging research prototypes to production-class ML/HPC platforms. Known for working at companies through strategic acquisitions, he brings both system-level architectural vision and hands-on microarchitectural expertise to optimize compute, memory, and interconnect for real workloads.
7 years of coding experience
23 years of employment as a software developer
MA ECE, MA ECE at University of Massachusetts Amherst
BA Physics, BA Physics at Tsinghua University
MA Physics, MA Physics at Brandeis University