Rami Cohen, PhD

I'm currently at Google, where I'm a senior AI systems architect focused on designing the hardware and software for next-generation AI accelerators. Before Google, I spent six years at Medtronic leading a team of data scientists, software engineers, QA engineers, and DevOps engineers, responsible for developing AI-powered computer vision applications for the healthcare industry. See my LinkedIn profile.

I graduated with a PhD degree from the Andrew and Erna Viterbi faculty of Electrical Engineering of the Technion - Israel Institute of Technology in 2017. In my PhD studies, I was advised by Prof. Yuval Cassuto, working on the design and analysis of error-correcting codes for high-speed memory devices and systems. Prior to that, I obtained my MSc degree from EE, Technion in the field of image/video processing, focusing on compression of medical videos, under the supervision of Prof. David Malah. I initiated and supervised more than 45 research projects in EE, Technion (topics include machine/deep learning, computer vision, signal processing and more). Click the Students & projects tab for additional details.

My research interests are broad:

·  LLMs and foundation models

·  Theory of deep learning

·  Image/video/audio processing

·  Image/video compression (e.g., H.264/H.265)

·  Error-correcting codes (new channel models, non-binary LDPC codes, efficient decoding algorithms)

·  Algorithms for the optimization of memory/storage devices with shared resources (e.g., cloud)