Now we’ve learned that the automaker is losing two of the chip architects behind the project.
Back in 2016, we first exclusively reported on Tesla quietly hiring legendary chip architect Jim Keller from AMD and we were fairly excited by the implications of Tesla hiring such an important chip architect.
At the time, we speculated that Tesla could be looking into making its own silicon at some point – speculation that was further reinforced after Keller’s hiring was followed by a team of chip architects and executives from AMD also joining Tesla.
Finally, our suspicions were confirmed two years later when Elon Musk confirmed that Tesla is working on its own AI chip.
Keller has since left Tesla, but the team is now led by Pete Bannon, who was amongst the other chip architects hired by Tesla over 2 years ago.
David Glasco, System Architecture Lead, and Dan Bailey, System Circuit Design Lead, were also among that group of engineers poached from AMD to work on the project.
But we’ve now learned that they are both leaving Tesla at the same time to take similar roles at Esperanto Technologies, which is developing computing solutions for AI/machine learning – not unlike what Tesla is trying to do with its Autopilot hardware 3.0 upgrade.
Esperanto announced their new hires today:
“Together, Glasco and Bailey will be responsible for the development, delivery and support of Esperanto’s computing solutions designed to achieve the highest levels of performance and energy-efficiency for artificial intelligence and machine learning applications. Esperanto is designing single chip SoC solutions using thousands of energy-efficient ET-Minion™ RISC-V cores and also high-performance ET-Maxion™ RISC-V cores in leading edge 7nm technology, based on the free and open RISC-V instruction set architecture.”
Glasco, who becomes Vice President of Engineering, commented on his move:
“My passion is developing the world’s best silicon products on the leading edge of technology, and AI/machine learning is the next and most exciting frontier. Within the next decade, the world will be very different as a result of innovation in these areas, and I am convinced that Esperanto and the open standard RISC-V architecture will be key drivers. I am excited to join the Esperanto team to build the future with products for the artificial intelligence era.”
Bailey, who becomes Senior Director of Engineering, also commented:
“I enjoy being part of, and helping to build, great technical teams attacking really tough problems. From what I know already, Esperanto has assembled a world class team and is working on solutions to energy efficiency and architecture challenges that will have a huge impact on the widespread adoption of machine learning solutions.”
We reached out to Tesla for a comment about the losses for their system architecture team, which is based in Austin. We will update if we get an answer.
Update: Tesla delicned to comment.
It sounds like Esperanto’s poaching effort resulted in a big loss for Tesla.
That said, it does sound like the bulk of the work on Tesla’s new AI chip has been done already and Tesla is mainly now trying to bring it to volume production in the coming months for a launch next year.
But that doesn’t mean that Bailey and Glasco wouldn’t have been useful at Tesla after the design of Autopilot 3.0 hardware is completed.
As I have been saying for a while now, computer power inside cars is becoming a highly critical feature and I think it’s important for automakers to get increasingly involved in chip making in order to be able to upgrade their onboard computer every year, like any other electronic device.
Therefore, I think Tesla’s AI hardware team is very important for the long-term success of the company.
Featured Image: A rare look at what Tesla Autopilot can see and interpret
Subscribe to Electrek on YouTube for exclusive videos and subscribe to the podcast.