Sihao Liu is a Research Scientist in the Architecture Research Group at NVIDIA Research. His work spans computer architecture, programmable accelerators, and AI-assisted chip design. He studies how to make specialized hardware easier to build and program through compiler–architecture co-design, automated design-space exploration, and agentic tools for electronic design automation (EDA).
His current research brings coding agents into architecture and EDA, combining implementation with independent review and executable evaluation. He created Humanize 1 at UCLA and contributes to Humanfia, which develops agent systems for long-horizon engineering tasks.
He received his Ph.D. in Computer Science from the University of California, Los Angeles in 2026, advised by Tony Nowatzki. His dissertation, Loom, develops a full-stack framework for compilation and exploration of heterogeneous, multi-core spatial accelerator systems. He also holds an M.S. in Computer Science from UCLA and a B.Eng. in Electrical Engineering from Xi'an Jiaotong University.
His research has received a MICRO 2022 Best Paper Runner-up award and IEEE Micro Top Picks recognition.
Research Interests
Computer architecture and programmable spatial accelerators
Compiler–architecture co-design and automated design-space exploration
AI agents for chip design and electronic design automation
FPGA architectures, networks-on-chip, and reconfigurable systems
