DataCamp is a good shout — I'll look at which track makes sense as a starting point. And I appreciate the honest answer on the 5-year question. "Who knows" is more useful to me than false certainty in either direction. The point about already having users is something I keep coming back to — it's the one thing that makes me think this isn't entirely wishful thinking. Did your wife find she needed to go through courses sequentially or did she jump to what was immediately useful?
Thanks — I'll take a look. Though I'd note the comment reads more like a product pitch than personal experience. If you've actually used it for something similar, I'd be curious what the limitations are.
I hear the frustration, and I think it's legitimate. You've put in years of formal work and the industry hasn't rewarded that fairly — that's a real problem worth being angry about, and it's not something I've had to face. I'm not going to pretend that building things with AI without formal knowledge is the same as what you've done. It isn't. The question your lawyer analogy raises is an honest one — professions that survive do so by protecting the value of deep expertise. I genuinely don't know how software settles that question as AI changes what "building" means. I'm figuring out where I fit in that picture, same as you seem to be. I hope you find what you're looking for.
That reframe is useful — "expert owner" versus "expert engineer" hadn't occurred to me quite so directly. I do want to understand the technical side well enough to not be dependent on anyone, AI or human, to make decisions I don't understand. But you're right that the ceiling on what I can build rises dramatically if I stop thinking I have to do everything myself. Worth thinking about where that line sits. Thanks for the perspective.