[mexicocitinluez]

I need you to understand that software development with LLMs isnt about writing a prompt to spit out your entire app.

[thefz]

I would not trust a single instruction that hasn't been reasoned over by a real person in my codebase. YMMV

[mexicocitinluez]

Why wouldn't there be a real person reasoning about it? Am I missing something?

[illiac786]

You are not, I believe the previous comment is not grasping what the different usage of LLMs in coding are. The previous poster believes LLMs are for vibe coding only, if I understand correctly.

[thefz]

No, I am not. I fully understand people use it on the full range from vibe coding to autocomplete on steroids. Trust none of these levels.

[illiac786]

I think the point we are trying to bring over is, we feel the way you wrote your statement implies you think no one is reviewing the output of LLMs and everyone is just blindly accepting it.

I disagree with this at least. Yes there’s a lot of people doing this, but I don’t think it’s the overwhelming majority, especially not when it comes to coding.

[thefz]

Nowehre I say anything about code reviews, my comment was about code generation

[mexicocitinluez]

Exactly.

[mexicocitinluez]

I don't know why this is all so difficult. It's been harder for me to convince software developers who think it's useless that there is value than to convince people who think it's a silver bullet that it isn't.