So far in my career I have always had more requests coming in than implementations going out. If I can go 3 or 10 times faster, than I will still have plenty of work. Especially for the slew of ideas that are never even considered to put towards a dev, because it's already considered to be too low value to have it even be considered to be build. Or the ideas that are so far fetched they were never considered feasible. I am not worried work will dry up.

What I believe is going to be interesting is what happens when non-engineers adopt building with agentic AI. Maybe 70 or 80% of their needs will be met without anyone else directly involved. My suspicion is that it will just create more work: making those generated apps work in a trustworthy manner, giving the agents more access to build context and make decisions, turning those one off generated apps into something maintainable, etc.

Or, there is just a lot more software written as the costs drop. I think most people work with software not tailored enough for their situation..