These things are in development for many years before they hit silicon, I don't think his time at intel has been long enough for that to be true. More likely, IMO, is that the same bureaucracy that lead to intel falling behind drives people like Keller away.

If scaling pipelines is hard, scaling pipeline pipelines is harder.

I would say that actually this is the time where you can have more freedom to get things done and be able take bigger risks.

Intel has to change something to catch up with AMD. So I imagine the 'play safe' people have less power now.

He was probably there specifically to help Intel regain that edge. But he may not have received enough support to implement his vision, or simply he had a personal reason to leave.