Even AI maximalists (of which I sure as hell am not) agree that the skill of the human in the loop (no matter the degree to which the human is plugged in) matters greatly for the output.

You still have to know things. You still have to learn things. How do you learn? Same as always. You learn by doing.

Being able to properly break down and understand a problem so it can be explained to the LLM is still valuable and something I think a lot of people who can already do it take for granted. Someone showed me their prompt about a project they were working on yesterday and it made no sense at all.

Being able to translate a problem into a solution AI can implement is still useful. I haven’t found AI to be good at designing systems or coming up with solutions to problems. It will try, but it always seems really bad. Having a person there to use AI as a tool, to guide it to solve the problem in an effective way, seems like the real skill.

What these tools look like in 2 years will likely be very different than what we see today. The main thing I’ve been doing is increasing my savings rate, and lowering my fixed expenses, so I have a lot of flexibility if/when it’s time for a new job or career. I know that no matter what happens, that will be helpful. Debt and any expensive fixed lifestyle costs represent risk. The more I can derisk, by keeping those values low and savings/investment high, the better.