Intel-Apple-AMD-Tesla Chip Architect Jim Keller Joins AI Startup Tenstorrent

Print Friendly, PDF & Email

Jim Keller, one of the leading names in HPC, AI and data center chip architecture, has landed at a Toronto hardware startup developing AI chips after a two-year stint at Intel that ended “for personal reasons,” according to the company, last June.

Keller has been named president, chief technology officer and a board member of Tenstorrent, which said in its announcement that Keller “will lead Tenstorrent’s efforts to be the hardware solution needed to address Software 2.0, the exciting industry shift towards using machine learning methods to solve problems previously addressed by traditional software.”

Keller’s career has included star turns at Apple, where he contributed to A-4-A7 iPhone and iPad chips; two stops at AMD to work on processors (including the AMD Zen, K7 and K8 x86 chips) that have enabled that company to mount competitive charges against Intel; and at Tesla to work on chips for autonomous vehicle systems as VP of autopilot and low voltage hardware. At Intel, Keller was SVP of the Silicon Engineering Group. In joining Tenstorrent, Keller is reunited with CEO Ljubisa Bajic, who had been a colleague of Keller’s at AMD.

Jim Keller

“Tenstorrent was founded on the belief that the ongoing shift towards ML-centric software necessitates a corresponding transformation in computational capabilities,” said Bajic. “There is nobody more capable of executing this vision than Jim Keller, a leader who is equally great at designing computers, cultures, and organizations. I am thrilled to be working with Jim and beyond excited about the possibilities our partnership unlocks.”

Last year, Tenstorrent, launched Grayskull, a 120-core programmable and scalable processor for machine learning architected, according to the company, to address the increasing size and complexity of AI models.

The company’s strategy fits into the larger trends of a) “technology disaggregation,” the big bang diversification of compute architectures away from Intel CPU dominance over the last 10 years; and b) the rise of chips designed specifically for AI and machine learning workloads, a phenomenon driven in large part by Nvidia GPUs, upon whom Tenstorrent is aiming is competitive sights, and further pushed by data center GPUs from AMD, AI processors from Intel and specialty AI chips such as from Cerebras, maker of the Wafer Scale Engine, which the company says is the world’s largest processor, with 1.2 trillion transistors and 400,000 compute cores.

“Software 2.0 is the largest opportunity for computing innovation in a long time,” Keller said. “Victory requires a comprehensive re-thinking of compute and low level software. Tenstorrent has made impressive progress, and with the most promising architecture out there, we are poised to become a next gen computing giant.”