How to Approach Learning in this Day and Age of AI?
I've always enjoyed learning about how systems work and how programmes operate but, in reality, I don't know syntax or specific file workflows.
I am still an amateur and all I really know how to do is write basic python with little or no libraries. The best thing I have every 'shipped' is a 2nd year winning python robot that used a competition custom built library that did all the action hardware for me. All we did in was plug in the motors and assigning 'R.motors[0] = 10' etc.
Not difficult, obviously. We just wrote navigation and logic.
I want to learn more and be able to make robots and programs but I don't know where to start with AI. Maybe 5 years ago, I'd know the route. Learn the syntax over months and build over years. But it's not like that anymore, it's not that simple because AI can do that anyway.
But it's also not "make this app" and then paste the errors into AI- that's vibecoding, right?
So far, with like 3 true but very unfinished projects under my belt, I have got AI to generate a tech stack based of inputs like 'this should be obtained from this API and then this should go to the frontend' rather than naming real libraries or packages because I simply don't know them.
People say, which I don't really believe is truly strong or sustainable advice, to get AI to generate it but then get it to explain it. When it generates an entire codebase, it can't really explain it within the size and window of a single reply, or even a full chat.
So, I suppose, the question I am asking is: Where do I actually learn in this day and age?
Thanks