Matt Welsh will help guide development of Xnor’s AI-on-the-edge platform
SEATTLE, Feb. 21, 2019 (GLOBE NEWSWIRE) — Xnor, the company that is making AI available on every device, has appointed Matt Welsh to the position of principal engineer. Welsh joins Xnor from Google, where he was an engineering lead on the Chrome Mobile team for the past eight years. Welsh’s mobile expertise will be crucial as Xnor scales its platform for consumer and commercial applications.
Xnor’s AI technology has already been adopted by companies in the security, home automation, industrial hardware, video conferencing and automotive industries. One of its computer vision models involving face identification can identify multiple faces quickly and accurately, directly on a device, making it ideal for home and building security systems and automakers. Unlike traditional AI, Xnor’s technology doesn’t require expensive hardware such as GPUs and neural accelerators. Xnor’s models are so compact and efficient they can run directly on a CPU with no loss in speed or accuracy.
Welsh’s experience solving a variety of difficult research and engineering problems made him a natural fit for the Xnor engineering team. “Matt has designed and implemented large systems that support more than a billion users,” says Jon Gelsey, CEO at Xnor. “His ability to think at a very high level about how Xnor works, from training through model deployment, will be critical as we scale our platform. We are excited to add his technical expertise and look forward to his leadership in making AI accessible on any device.”
Prior to Google, Welsh was a professor of computer science at Harvard, where his research focused on systems, networks and wireless. Welsh also taught the operating system class at Harvard in which Mark Zuckerberg was a student. He is the author of several books about the Linux operating system.
“Xnor is doing work on the bleeding edge of AI and embedded systems, both areas I’m passionate about,” says Welsh. “There’s a tremendous opportunity to make this technology more widely available, and this is the perfect next chapter for me professionally.”
Xnor is dedicated to accelerating AI and deep learning in consumer and business devices. Its enterprise AI models and Ai2GO self-serve developer platform deliver AI everywhere, for everyone. Unlike traditional AI that either runs in the cloud on massive data centers or requires expensive hardware such as GPUs or neural accelerators, Xnor enables AI to run at the edge on inexpensive devices with no internet connectivity. This new, highly scalable approach ensures complete privacy of data, eliminates the need for connectivity, and significantly reduces memory load, latency and power consumption without compromising accuracy. Xnor was built using award winning research conducted at the Allen Institute for Artificial Intelligence. It is used by global corporations in aerospace, automotive, industrial hardware, retail, photography, and consumer electronics. Founded in 2017, Xnor is headquartered in Seattle, Washington.