TLDR: Scaling AI @ Tesla, ex-Googler github.com/zhuangh, linkedin.com/in/zhuangh/
Prior to that, I solved problems ranging from software to hardware and led a team at Google, mainly worked on efficient machine learning compute for media accelerators, AI perception, video understanding, ML-assisted video compression, and contributed to Youtube Datacenter VCU and Google Tensor chip. My former team at Google also maintained software package QKeras (co-invented with Dr. Claudionor Coelho) for ML software/hardware co-design, which is also supported by other open-sourced software toolchains to convert quantized ML models to hardware implementations such as FPGA and ASIC. QKeras was used in our Nature journal (CERN News).
I got my Ph.D. in Computer Science from UCSD cs.ucsd.edu/~hazhuang/. During the acdemic year, I numerically modeled and analyzed dynamic systems inside large-scale integrated circuits (VLSI chips) with the help of matrix computations from high-performance computing area. I also did a little bit of place-and-route in VLSI physical design. My main research from master's and PhD all got commericalized and transferred into the EDA industry. I spent my last year of my PhD study in industry and did those technology transferring.
Copyright © 2023 FlameSync LLC. All rights reserved.
