I think the main barriers are context length (useful context. GPT-4o has “128k context” but it’s mostly sensitive to the beginning and end of the context and blurry in the middle. This is consistent with other LLMs), and just data not really existing. How many large scale, well written, well maintained projects are really out there? Orders of magnitude less than there are examples of “how to split a string in bash” or “how to set up validation in spring boot”. We might “get there”, but it’ll take a whole lot of well written projects first, written by real humans, maybe with the help of AI here and there. Unless, that is, we build it with the ability to somehow learn and understand faster than humans.

People seem to disagree but I like this. This is AI code used responsibly. You’re using it to do more, without outsourcing all your work to it and you’re actively still trying to learn as you go. You may not be “good at coding” right now but with that mindset you’ll progress fast.

Not what I’d have expected. In my company it’s mostly higher ups (suits) pushing the stuff and workers begrudgingly implementing it.

deleted by creator

How high up in the corporate ladder are they?

As a former script kiddie myself I think it’s not much different from how I used to blindly copy and paste code snippets from tutorials. Well, environmental impact aside. Those who have the drive and genuine interest will actually come to learn things properly. Those who don’t should stay tf out of production code, which is why we genuinely shouldn’t let “vibe coding” be legitimized.