I started work in systems and devices for optical computing at Caltech, developing neural networks and radar signal processing systems. These analog systems were the first large scale adaptive neural networks, utilizing volume holographic adaptive weights for highly parallel systems. I founded Network Photonics, developing WDM digital communication systems for metro area optical networks. I am also a member of the graduate faculty at the University of Colorado, Boulder and am on the editorial board of Fiber and Integrated Optics, the first academic journal in the field.
I was inspired by the vision and goals of Fathom and joined for the opportunity to develop the most powerful optoelectronic computing system. My formal background is in innovative computing architectures, starting with classes taught by Richard Feynman, proceeding with neural networks and analog optical processing and culminating in digital communications. This background leads directly to the innovations Fathom is developing.
Fathom’s architectural philosophy is to design for ambitious systems without taking on unnecessary risk. Our team has a deep background in traditional and alternative computing technologies and we are using that experience to target a system with modest algorithm and execution risk while pushing for maximum performance advantage. I am excited to contribute my skills and experience to the team and to be part of what promises to be a revolutionary step in computing.