LLMs turn your job into mostly code review, a task everyone famously loves to do and is good at
@jcoglan
There is no such thing as code review.
@janneke not sure I understand what you mean by this, can you elaborate?
@jcoglan
I'm also interested in learning more. In some metaphoric kind of Un-Code-Review way, as with unconferences, or as a matter of fact?
@janneke

@yala @jcoglan
When I learnt programming in the 80s, I would always do that together with a friend. Having access to a computer was hard to get and you would share that time. After a brief stint as a lonely "professional" programmer in the 90s, that was "allowed" again but you had to call it pair programming.

I've been trying to avoid the ritual commonly known as code review (trckacr) for many reasons. I don't like it as a reviewer. I'm bad at it. I'm not a computer and something that looks plausible to me may not even compile. Cosmetic or stylistic errors/anomalies are easy, but reviews usually happen when the code is finished. Are you really going to suggest a full rewrite? A good friend of mine was a team lead at Google, and they lamented about how impossible code review sessions were and how producing quality code this way seems impossible.

It's not all black and white. I do find review comments helpful sometimes, especially when entering a new project. Usually that's about style or other cultural memes.

If you want to have (give or receive) truly impactful input from a second person on a piece of code, pair program it.

Anyway, I believe that the person who codes alone as opposed to pairing is cultivating the software crisis, and so is trckacr.

@yala @jcoglan
So yeah, I do acknowledge the ritual exists, just not that it has much if anything to do with what it suggests that it is, or with what managers (non-programmers) belief it is.