This tweet from Andrej Karpathy[1] coined the term and provides a good definition (with a few snips for brevity):
> There's a new kind of coding I call "vibe coding", where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It's possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good. ... I "Accept All" always ... When I get error messages I just copy paste them in with no comment, usually that fixes it. The code grows beyond my usual comprehension, I'd have to really read through it for a while. Sometimes the LLMs can't fix a bug so I just work around it or ask for random changes until it goes away. ... but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.
It's not that hard to understand that this is a distinct thing from "AI-assisted" development. It's a distinct thing that can be fun and no serious person thinks it should be your only modality for getting real work done.
If you've got an AI-aided workflow and uninformed people are dismissing what you're doing as "vibe coding" just... idk, ignore them? My opinion is that anyone who hasn't at least tried this stuff and taken the time to understand how it could be useful to them is just not worth listening to.
Also,
>Real devs use printf, not debuggers.
>TypeScript is just training wheels.
Ironically using a debugger over print statements and TS over JS requires more effort.
truelinux1•58m ago
I wrote this because the term “vibe coding” irritates me. This meme has taken on a weird, dismissive tone that doesn’t reflect what serious AI-assisted programming actually feels like.
The piece is about how this meme obscures a genuine shift in developer workflows, especially for the many people who are actually shipping production code with aid of LLMs.
Curious how this community sees the term. Is it helpful? Misleading? Harmless? Something else?
KerrAvon•13m ago
No, “Vibe coding” is a pretty accurate descriptor for the AI slop code dumps a lot of people are putting into PRs now. Maybe it’s acceptable for web UI front ends, but anything people actually rely on needs to be held to a higher standard.
Very general rule-of-thumb: Doesn’t matter if the AI wrote some or all of the code. If you, personally, can explain every line in the PR, and justify its existence, it’s fine; it’s not vibe coding.