LLM + framework you don't understand goes in ... unmaintainable garbage comes out.

▲ CSSer 3 hours ago | parent | next [-]

Before LLMs, learning on the job looked like reading documentation. Now it’s a guided tour with verification. When I produce things in this way, I’m not just blindly accepting it. The goal is that by the end of it I have learned more about the codebase and architecture, not less. I feel that’s important.

▲ skellera 2 hours ago | parent [-]

Many people don't understand this, even big tech engineers. They see LLMs as a bottleneck. It's more that they don't understand how to use it to multiply their skills, just basics and code gen.

	▲	hungryhobbit 2 hours ago \| parent [-]
		I use multiple Claudes at a time, daily. It's precisely because of that experience that I wrote: `LLM + framework you don't understand goes in ... unmaintainable garbage comes out.` Claude follows code patterns and structure. If you setup that structure and those patterns properly, it will produce great code. If not, it will follow ... whatever it feels like, with each commit. If you just have it built something with a framework you don't understand, it will do so just fine! But over time every "vibe coded" change you make will drift it further and further, until you are left with a mess of vibe-coded spaghetti.

▲ 6510 2 hours ago | parent | prev [-]

Using it to understand a framework is fine.

▲

repelsteeltje 2 hours ago | parent [-]

I agree that's fine in that at least it's doesn't cause unmaintainable garbage. And might even get you up to speed quicker that reading the docs old school.

But the GP point, that you're better off finding people that already, truly understand and are familiar with the tech (ie. Astro), imo still stands.

	▲	6510 an hour ago \| parent [-]
		We cant all be wise enough to use php. I read a fun comment the other day from a frustrated windows user who failed to configure linux in previous attempts but now with LLMs it was very easy for him.