[blihp]

Obviously not the only reason, but he appears to be a helpful ingredient in the mix with good things tending to happen with him around. No idea how much of it is due to his technical vs. management skill, but there definitely seems to be some there there.

[ylk1]

He was an actual Microarchitect in his early career. This guy clearly has good knowledge on every aspect of microprocessor design: transistor, interconnect, process complexity, core layout & utilization, ease of design, circuit complexity, time to market, micro architectural behaviors, ISA & software patterns. From what I've seen he is really good at setting targets for the teams, especially given that he has been at multiple companies, he clearly knows what is good and what is lacking. Setting aggressive targets and motivating the team towards them. Something the previous management failed at.

[patrec]

In that light AMD losing him to Intel via Tesla sounds like a potential few billion dollar blunder. I wonder how difficult/expensive it would have been to prevent that.

[jessermeyer]

Jim has a long history of jumping between companies.

[patrec]

All the more reason to try hard to avoid him ending up at your main competitor if you finally start getting a leg up.

[adventured]

You probably can't. He appears to like jumping into challenges such as those that AMD was facing and Intel is facing now. The jumping would seem to imply he's doing it to pursue an interesting challenge as the primary. Some personalities need that (others are obviously repelled by that risk & challenge). Lots of big corporations can afford to pay very large sums to retain him, if that were his primary consideration.

[hanniabu]

Ask HN: I'm somebody with this type of personality that likes challenges and to bounce around, but I'm not as talented and experiences as someone like Jim. What are some suggestions on how I can satisfy this while also avoiding the trap of not being somewhere long enough to gain a higher level of experience? It seems that my habits have sabotaged me and my peers have excelled in their companies to much higher positions and compensation.

[greggyb]

Feel free to email me to talk more about this. I can share trajectories for myself and a few peers who have similar experience.

Get a role at a >decent consulting firm in the field you like. Move aggressively into a role where you get to take lead on projects. Find conferences in your area of interest. Speak at those. Build relationships with any vendors that are common to your customers and area of interest. Speak more. Get to know organizers at conferences and seek keynote and panel discussion opportunities. Engage with everyone you can to identify problems. Evolve your content to focus on the intersection of interesting and common problems. Somewhere in here you can shift to a top-tier consulting firm in your field, or to independent consulting, or jump to a vendor, or take on a senior role at an org that would otherwise be a customer of your consulting firm.

At this point you should have strong experience and a reputation for the same. Leverage this to filter opportunities to those you want.

All of this is predicated on you actually being quite good at what you're interested in. You don't have to be world class to start, but you do need to continuously improve. You'll probably end up in the top 10% of your field. Again, predicated on ability.

[jessermeyer]

Jim has extremely strong foundations in his field of study, and is work is an outpouring from this. He bounces between companies, but not from the basic problem of CPU arch and management, allowing each problem to further deepen his roots.

[adventured]

That really depends on the specific nature of your bounce around trait, what you need to fulfill that part of your personality. Only you can properly answer that of course. However, a few speculative ideas:

- Consulting can be interesting to feed that. You can expose yourself to a wider variety of projects and you can eventually have a lot of say in what work you choose to take on. You will also have the opportunity to limit the duration of projects you take on to an extent, so you won't be stuck on any given thing for longer than you can tolerate. Difficult to get started, to gain momentum.

- A first line suggestion from HN would typically be side projects. This can help restrain your desire to bounce around, dulling that like a pain killer. You do what you want on your own time, and change it up anytime you see fit, while staying at a job and trying to progress up the ranks. I don't know what your skillset is of course (eg programmer), it's somehwat dependent on that as to whether side projects is an interesting angle.

- If you think you're not experienced enough, focus on trying to leap yourself forward on something you are good at, to open up more opportunities to jump. Push one of your skills well above the market average. If you have some strong center pivot skill, you can bounce more often. You don't have to be Jim Keller to do this, people that are in the top 1/3 in tech at something often bounce around because they can, it's not unusual. As the other person commented, Jim has a core strength and it's why he can jump around. The bottom 1/3 is in a beggar position, they are always in a position of having to take what they can get by necessity (which rarely changes, unless their skill level changes or the labor market is extremely tight); so you have to focus on pushing a skill as high as you can, and you don't need to reach elite levels as an outcome.

- Much like physical exercise (high intensity training), there are paths to gain intense experience in shorter amounts of time with a high payoff in the experience you acquire. These opportunities are rare, although that doesn't mean they don't exist. The work is usually very difficult, the hours are more likely to be long. There is always that trade-off in there. To get what you want in the bounce around aspect, you might have to dial up the intensity for a time until you get the experience you need (after which you can dial it back).

[p1necone]

Probably a good thing for the market at large though. If the current x86 duopoly ever turned into a monopoly with Intel or AMD it would not be good for consumers.