[rfgplk]

> the result is some massive crap in nextjs that needs 10GB mem to compile, has 1000s of lint errors, dev logs in git (very noisy ones) and so on.

The anti-LLM propaganda is getting ridiculous at this point. No project "needs 10GB" to compile, unless you're working with astronomically massive repos, and _no_ LLM will _ever_ generate that. Lint errors (depending on cause) are either meaningless or a result of poor prompt engineering. If you want your project linted/formatted a certain way make it clear to the LLM.

[anonzzzies]

Well, this is not the first project that has this. Depending what llm/harness and how it is yielded, this happens, a lot. How often have you had Opus 4.x note; ‘485 failed tests, but those are not related to my changes so I will not dive into them’? You believe someone who does not know about tests or code or linting or compilers will push back on that? I know that’s not happening for a fact.

[boringg]

>If you want your project linted/formatted a certain way make it clear to the LLM.

Most people have no clue what that means. Most Devs should know but I would wager that newer devs don't have enough opinion or exposure to do that.

[sensanaty]

I have approx 40GB of docker containers spun up at any time in order to run the app I work on, and it only seems to get hungrier by the day. I have absolutely 0 problems imagining a single unoptimized Next app needing multiple gigs just to run, yet alone compile.

[hightrix]

Linting is also trivial for LLMs. I’d argue linting is one of the easiest and best use cases for coding assistants.

This story doesn’t add up.

[tayo42]

The person running the Ai needs to know to set up linting in the first place and have it be applied. It's not a default.

Ive had ai make code that doesn't pass a linter and make code look like hand aligned code.

[aleqs]

Shameless plug - I've been working on a general repo linting tool alint [0], which helps keep a repo cleanly structured and hygienic - essentially a linter for everything in a repo that a language-specific linter doesn't cover. It also has some considerations for integrating/playing nicely with agentic coding [1]. This started as a replacement to a bunch of scripts I had in repos that did similar things - but more cohesive, readable and much faster.

[0] https://github.com/asamarts/alint

[1] https://alint.org/agent-friendly-linter/

[nozzlegear]

You'd have to know what a linter is and tell the LLM to set it up though, right? Or just get lucky and have it suggest setting it up/set one up without asking.

[mannanj]

The OP said that this was coded by a product person though. The product people coding and using these tools, who are spreading the pro-LLM propaganda, are not doing proper prompt engineering nor lint fixing.

> _no_ LLM will _ever_ generate that

Did you even read the post? Seems you are being overly hostile, defensive and dismissive. Honestly, you sound like an astroturfer to me. I'm curious to check your history to see if you match the vibe.

[hylaride]

Oh sweet summer child...

Do you have any idea, even before modern LLMs, how much code out there was just given more memory instead of optimized? There are even good reasons for it (it can be cheaper than having devs spend time if there's more productive work to be had).

It can be very easy if said code needs to parse through gobs of data.

[thr0w]

> No project "needs 10GB" to compile

You've never tried to compile NextJS slop, have you? It absolutely can take that much. All those junkdevs have 64GB in their MBPs for a reason. NodeJS max heap is now dynamic because of this.