Talent Wars

Posted: 06.29.18

The revolving doors in Silicon Valley are nothing new, but within the last two or so years the migration at the superstar techno level has been particularly intense—especially in the GPU and CPU space.

Some recent examples:

Martin Ashton served as EVP of PowerVR Multimedia IP at Imagination Technologies Group Plc until December 1, 2016 where he was instrumental in developing PowerVR graphics technology. He then moved to Intel, and most recently has moved on to AMD.

Jim Keller was corporate VP and chief cores architect at AMD, where he led the development of the Zen architecture. Previously, Keller was vice president of Engineering and chief architect at P.A. Semi, which was acquired by Apple Inc. in 2008. He led Apple’s custom low-power mobile chip efforts with the original A4 processor that powered the iPhone 4, as well as the subsequent A5 processor. In April he joined Intel to lead the company’s silicon engineering, which encompasses system-on-chip (SoC) development

Raja Koduri  was at AMD, where he was senior VP and chief architect of the Radeon Technologies Group. In this role, he was responsible for overseeing all aspects of graphics technologies used in AMD’s APU, discrete GPU, semi-custom and GPU compute products. Prior to AMD, Koduri served as director of graphics architecture at Apple. In November 2017, he left AMD and went to Intel to be chief architect, senior vice president of the newly formed Core and Visual Computing Group, and general manager.

Tim Leland VP of products at Qualcomm and driving force behind their GPU and Snapdragon development left to be the Vice President / GM of Frostbite Engine at Electronic Arts. Some of his Qualcomm GPU team left as well to go to AMD and Cisco.

John Metcalfe, Imagination Technologies COO left the company for Apple, and is now listed as a Senior Director at Apple. Key technologist Jonathan Redshaw followed him there a little later.

David Wang rejoined AMD as senior vice president of engineering for Radeon Technology Group (RTG). Prior to that he had been at Synaptics, where he was senior vice president of Systems Silicon Engineering. Prior to joining Synaptics, Wang was corporate vice president at AMD responsible for SOC development of AMD processor products, including GPUs, CPUs, and APUs. He left AMD when VP Rick Bergman left the company for Synaptics. (Sort of ancient news, but it’s all part of the same carnival ride.)

These are not gig-workers, tinkerers going from camp to camp, they are mid-aged and older guys and gals making serious career and potentially life changing moves, so it’s not done lightly or perfidiously. It’s not just about money, although that’s certainly a factor. And some mighty stellar salaries and stock options are being offered.

This apparent nomadic behavior isn’t going to abate because in addition to being attracted by money, there is also the bigger desire to do something grand, to design something new and amazing. The mystic and superstar status legends created for and sometimes by these movers and shaker, who are literally movers, add to the excitement and rush, and fill the social media channels.

As noted in previous essays and editorials, it takes three to five years from inception to realization (see Raja Koduri to lead Intel to the promised land AI, AMD at Intel) for a GPU or CPU, and often the designs are multi-authored with the initial architect leaving and new person coming in. The designs become blended and homogeneous with the IP boundaries grayed out and unassailable—not that some ambitious paid by the hour lawyers won’t try to assert infractions and IP theft. How does a brilliant designer or engineer forget things he or she has learned and seen at company A when he or she moves to company B? You can’t. And the law says you can’t be denied employment for what you know, so the migrations aren’t going to end.

People whose livelihood depends on picking winners and losers both short term and long term watch these affiliation shifts Then, they try to crystal-ball a scenario that predicts what it means. The net result is a superstar that joins company B and is given a stock option can often enhance his or her profit by having the stock pickers run the share price due to the movement to company B.

Team mates often go with the superstars. Another reason people move from one company to another is because something has changed (or hasn’t changed enough) in company A. When it’s no longer fun, or potentially rewarding to be at company A, the most valuable people leave, and often their colleagues follow them. That complicates the equation because there are laws against poaching. So the second-tier or adjunct people have to be careful when they move, as do the hiring companies.

Another aspect to these employment changes is often a promotion, and that adds an element of risk for the new employer. It’s the candidate’s history that makes him or her attractive, but can they do something they haven’t done before? It’s not like moving from the right seat to the left seat in cockpit—co-pilots mostly fly the planes anyway. But can a really good circuit designer become a really good architect? Can an architect become a GM?

These movements from company to company aren’t new, but they have become more intense and more high-profile.

Attracting and keeping talented engineers has always been a challenge, and it feels horrible when one of them leaves, especially if to a competitor. The top management, if they have any sense asks what could/should we have done differently. And then if they find an answer, can they, will they do it so more IP doesn’t walk out the door, and/or they can attract new talent? Usually talented engineers and marketing people leave, or get fired when a company is acquired. It makes you wonder what the acquiring company thought they saw in the company before they acquired it, and what went wrong afterwards. You don’t hear too much about those displaced souls, or the folks who joined a startup that failed to start, or sustain itself.