Home

Harvard Mixed-Signal VLSI

Combining Circuits and Architecture to Combat Variability

Gu-Yeon Wei
Associate Professor of EE
Harvard School of Engineering and Applied Sciences

I received my BS, MS, and PhD degrees in Electrical Engineering all from Stanford University in 1994, 1997, and 2001, respectively. In August 2000, I joined Accelerant Networks (now a part of Synopsys) in Beaverton, Oregon as a Senior Design Engineer, where I worked on a 5-Gbps backplane transceiver with adaptive equalization. As the internet bubble was showing signs of bursting, I joined Harvard University as an Assistant Professor of Electrical Engineering in January 2002. I was promoted to Associate Professor in 2006 and currently awaiting my tenure case decision.

I started my research career thinking about system-level energy optimization and my PhD thesis focused on implementations of power-supply regulators for dynamic voltage and frequency scaling (DVFS) and how it can be applied to high-speed parallel links. Since then, my research interests have broadened somewhat. While I still work on a variety of links-related projects, I’ve also been collaborating closely with Prof. David Brooks on several projects to investigate different ways circuits and architecture (and software when appropriate) can work together to better deal with variability issues that affect modern IC designs. Several of us also recently received $10M funding through the NSF Expeditions in Computing Program to build RoboBees. I will be working on the power electronics for the ‘body’ and circuits + architecture for the ‘brain’. Additional information on my research activities can be found on the MSVLSI group website.

Besides my research activities, I teach students how to design chips in the VLSI Design course (CS148/CS248) and think about important circuits and systems issues through a graduate-level reading course that covers a wide range of topics related to Mixed-Signal IC Design (ES271r).