I have a sneaking suspicion that if everyone proffering an LLM coding opinion stated their languages and area of focus/expertise we’d all be in a lot more agreement about the output…

Example 1: “as a JavaScript front-end developer working with React and node, I think LLMs save 500 hours a week and are better than anyone on our team.” <— easy to believe

Example 2: “as a backend Erlang developer working on life-critical systems the inconsistent output makes LLMs dangerous and the code is tidy but terrible at scale.” <— fully congruent with the first example

Mm.

As an iOS developer, I have yet to see two different teams agree on best practices, architecture, or code style, and the only reason I've seen anyone agree on UI frameworks is that I've been in slightly more teams/seen more ads than there even are widely-used UI frameworks; and despite all the big promises in the job adverts, 95% of the work has been to take someone else's existing UI design and connect it to someone else's existing API.

LLMs fit right in.

(Unlike me: I'm burned out by iOS and LLMs came at exactly the right time).

Yeah, that's what I think every time I read some discussion online about AI coding assistants. Programming is such a wide field with presence in so many different industries, with all sorts of languages and environments, and many people seem to forget that not everything is webdev. You see one person saying they barely need to code by hand anymore and another who can't believe that because their experience has been the complete opposite, when they're probably both right.

Even Example 1 works in a much narrower scope, in that you need to have a decent working setup and all the rails like a good AGENTS file and so on.

The point is that "Vibing" still does not produce worthy output.