Less than 2 percent of all cars sold in America today are electric, but that will soon change. The internal combustion engine’s firm grip on American transportation is loosening and for the first time... Read more about Prepare the grid now to power electric vehicles
Mohamed Abdelfattah: Designing the next generation of computers
- New Faculty Year: 2022
Assistant professor, Electrical and Computer Engineering at Cornell Tech
Academic focus: Machine-learning-centric computer systems.
Research summary: My research lies at the intersection of computer hardware and machine learning. I investigate methods to codesign algorithms, software and hardware to optimize the efficiency of emerging applications powered by machine learning. On the hardware side, I focus on reconfigurable computing with devices like FPGAs, especially in the datacenter setting. On the algorithms and software side, I have been mostly working on automated machine learning (AutoML) methods to enable neural networks on constrained mobile devices. My overarching goal is to design the next generation of computers that are powerful yet efficient to power machine learning applications.
What inspired you to pursue a career in this field? I was always fascinated by electronics. Once I completely tore apart one of my toys, I would end up staring at this thin black plastic casing that envelopes an integrated circuit and I knew that it was the brain behind the whole device. During my undergrad, I was eager to understand how these integrated circuits worked. I took many courses on semiconductor technology and analog electronics but then quickly fell in love with digital electronics, and particularly FPGAs. I researched FPGA devices throughout both my master's and PhD, only to confirm my passion for hardware acceleration and spatial computation. More recently, machine learning has completely captured my imagination but further progress is literally blocked by the availability of compute resources. This is why I am very much motivated to work on the next generation of computers, to enable further progress in machine learning.
What are you most looking forward to as a Cornell Engineering faculty member? Having worked in industry (Intel and Samsung) for the past 6 years, I am excited about working with students again in an academic institution. I am looking forward to creating multiple collaborations with both academic researchers and industrial partners, and I hope to take full advantage of open-ended academic research in pursuing forward looking research ideas that are not necessarily tied to short-term industrial goals. At Cornell, I am really looking forward to being a part of both the Computer Systems Lab and the Tech campus in NYC, both with an amazing roster of world-class researchers.
What do you like to do when you’re not working? I like to explore the world with my wife and two children, whether we're cycling around our neighborhood, or camping in the middle of a desert. I also regularly go swimming and occasionally play squash.