Vibe Coding is not an excuse for low-quality work • Hacker News Board

Re: Vibe Coding is not an excuse for low-quality work

#201

Post by QuadmasterXLII » Mon, Apr 21, 2025, 1:30 AM UTC

simonw wrote: post #51

SCdF wrote: post #5

So what have we redefined vibe coding to mean exactly? The original tweet[1] talked very specifically about not caring about quality, just accepting whatever code the AI produces blindly, as long as you get the black box output you're looking for, and just randomly try again if you didn't. Are people now using this term to mean "giving an AI agent broad tasks"? [1] https://x.com/karpathy/status/1886192184808149383?la…

I wrote about this last month: "Not all AI-assisted programming is vibe coding" - https://simonwillison.net/2025/Mar/19/vibe-coding/ Vibe coding is when you don't review the code at all. If you're using LLMs to help you write code but you're actually reviewing what they produce (and iterating on it) that's not vibe coding any more. This battle is almost certainly lost already, but dammit I'm gonna keep fighting anywa…

Vibe coding is shifting to mean when they don’t review the code, vs the original meaning of when they don’t intend to review the code.

Re: Vibe Coding is not an excuse for low-quality work

#202

Post by Capricorn2481 » Mon, Apr 21, 2025, 4:57 AM UTC

dheera wrote: post #118

simonw wrote: post #52

Earlier quoted context omitted.

"The whole point of AI agents is to eventually get good enough to do this stuff better than humans do" You can be an enthusiastic adopter of AI tooling (like I am) without wanting them to eventually be better than humans at everything. I'm very much still in the "augment, don't replace" camp when it comes to AI tooling.

No, I actually do want them to eventually be better than humans at everything. I'm okay with directing AI to write exactly the software I need, without having to manage "stakeholders", deal with the egos of boomers up the management chain, and all the shitty things certain people do in any organizations that get large enough.

Why do you think any of those are going away? You can be the best programmer in the world and you'll still be dealing with that.

If you're in a situation where you don't have to deal with that, it won't be the result of using AI

Re: Vibe Coding is not an excuse for low-quality work

#203

Post by namaria » Mon, Apr 21, 2025, 9:01 AM UTC

mulmen wrote: post #188

otabdeveloper4 wrote: post #153

Earlier quoted context omitted.

> yeah bro, leave the easy and fun part (googling and typing code) to an AI, and let the human deal with the tedious and error-prone part (checking all the edge cases) What did you really expect?

Huh, I have never found searching for information and typing code particularly fun. The part I enjoy is seeing something work.

What I personally like is solving a problem in my mind and then building a solution.

Watching a solution being built after describing the problem robs me of that joy. And of the growth opportunity in flexing my reasoning.

Re: Vibe Coding is not an excuse for low-quality work

#204

Post by michaelbrave » Mon, Apr 21, 2025, 10:22 AM UTC

jhbadger wrote: post #182

SCdF wrote: post #5

So what have we redefined vibe coding to mean exactly? The original tweet[1] talked very specifically about not caring about quality, just accepting whatever code the AI produces blindly, as long as you get the black box output you're looking for, and just randomly try again if you didn't. Are people now using this term to mean "giving an AI agent broad tasks"? [1] https://x.com/karpathy/status/1886192184808149383?la…

It's like doomscrolling. Originally it meant compulsively reading depressing content about the state of the world, wars, etc., explaining the "doom" part, but eventually it evolved (or perhaps devolved) into just meaning obsessively checking any sort of online content whether or not it was depressing.

in a lot of ways that is understandable when so much of the news is bad news, if all the highlights of the day point toward the world going to hell even reading a normal newsfeed can feel like doomscrolling.

Re: Vibe Coding is not an excuse for low-quality work

#205

Post by dbbk » Mon, Apr 21, 2025, 12:26 PM UTC

closewith wrote: post #170

dbbk wrote: post #169

Earlier quoted context omitted.

LLMs don’t even understand fucking TypeScript, which you would expect a computer program to be able to understand. I can’t get it to write valid Drizzle code to save my life, it will confidently hallucinate method imports that don’t even exist.

The question is which LLM, invoked how? Claude Code refactored numerous projects for us into TS, often one-shotting it. Saying LLMs don't understand TS (which may be true, in that LLMs questionably understand anything) says more about your perception than model and agent abilities.

Right, LLMs don't understand TS, because they're not integrated with it. When they come across something they don't know, they just start hallucinating, and don't even verify if it's actually valid (because they can't)

Re: Vibe Coding is not an excuse for low-quality work

#206

Post by closewith » Mon, Apr 21, 2025, 12:38 PM UTC

dbbk wrote: post #205

closewith wrote: post #170

Earlier quoted context omitted.

The question is which LLM, invoked how? Claude Code refactored numerous projects for us into TS, often one-shotting it. Saying LLMs don't understand TS (which may be true, in that LLMs questionably understand anything) says more about your perception than model and agent abilities.

Right, LLMs don't understand TS, because they're not integrated with it. When they come across something they don't know, they just start hallucinating, and don't even verify if it's actually valid (because they can't)

LLMs can't, but agents can. They can read documentation into context, verify code, compile, use analysis tools, and run tests.

Hallucinations do occur, but they're becoming more rare (especially if you prompt to the strengths of the model and provide context) and tests catch them.

Re: Vibe Coding is not an excuse for low-quality work

#207

Post by CRConrad » Mon, Apr 21, 2025, 7:34 PM UTC

Funny, that's all I thought it ever was.

Re: Vibe Coding is not an excuse for low-quality work

#208

Post by marcusb » Mon, Apr 21, 2025, 9:10 PM UTC

t-writescode wrote: post #113

marcusb wrote: post #20

Earlier quoted context omitted.

The definition seems to have pretty rapidly moved to 'used an AI coding assistant in some capacity'.

And I think that *horribly* muddies the definition because programming and getting a bit of AI help here and there (since it's read the docs better than you have, and scanned Stack Overflow for common problems and mistakes more thoroughly than you have) is, imo, a very, very valid way to program; but you're still very much in the driver's seat. It's not creating whole design patterns, etc, for you. I wonder if maybe…

I agree - I'm not endorsing this broad usage of the term, just noting how I see it used in practice. This seems to be half driven by grifters on social media promoting their AI coding course du jour and half by people who just discovered Cursor, Windsurf, etc., and think any use of those tools counts.)

Another commenter said these people only read half of Karpathy's tweet - I disagree. I don't think they read it at all, have heard the phrase, and are using it with reckless abandon.

Re: Vibe Coding is not an excuse for low-quality work

#209

Post by freddealmeida » Tue, Apr 22, 2025, 1:31 AM UTC

Vibe coding for me meant a roll-the-dice approach to building something that world. Never a care for strong architecture or good code standards. Allowing anyone to become an "engineer". A friend of mine, who can't code, used Cursor to build fully functional Nextjs web apps. I was impressed. Vibe coding is a super power for average people.

I do think this article fully grasps that change. Using AI to do tasks is not vibe coding. At least to me.

When will the first vibe coding book come out I wonder?

Re: Vibe Coding is not an excuse for low-quality work

#210

Post by schwartzworld » Tue, Apr 22, 2025, 2:57 AM UTC

ehnto wrote: post #147

Workaccount2 wrote: post #136

Earlier quoted context omitted.

AI is to SWEs what self-driving cars (or like features on new cars) are to taxi drivers. A neat thing that may make some basic things a little easier. But it's the wrong perspective. Think instead what a self driving car is to someone who cannot drive. It doesn't matter if it sometimes disengages to have an operator step in or can only handle basic local trips. It's a total game changer. AI is unlocking computers for…

> AI is unlocking computers for people who have never written a line of code in their life. I don't think I can dispute the claim, but it feels more like someone who can't build a house being able to do it now that they have YouTube tutorials. Unless the person was already quite smart and competent, the house will probably have significant structural issues. Is that a bad thing? For housing, governments the world ove…

I couldn’t build a house, but I did learn a lot of smaller skills I would have otherwise called a pro in over. Turns out changing a toilet is pretty much putting it in place and screwing on the pipe, and you can program keys to my Xterra by toggling certain controls in a certain order.

I wouldn’t expect a vibe coder to build a full featured app on vibes alone and produce a quality codebase, but for smaller tasks and apps it is just fine.