Benjamin Lefaudeux 🇺🇦 @bentheegg.bsky.social · 18/07/2025In the coming age of agents, I think vibe coding will die out, same lasting power as prompt engineering. For things LLMs excell at, you might as well stick to higher level directives and let it own the work, Claude Code is a good example. 1/2 341
Benjamin Lefaudeux 🇺🇦 @bentheegg.bsky.social · 18/07/2025For things LLMs are not great at (typically new, frontier work) you're better off doing it instead of inheriting a broken spaghetti plate. Vibe coding your way to oblivion is not a great proposition for either of these. I don't think there's that much of a middle ground 110
Someone @aneuroticsomeone.bsky.social · 18/07/2025The problem with LLMs (and any other large neural network, really) is that too much hope is placed on emergent behavior. Some form of knowledge injection is necessary, not even humans can learn with a "here's what I want, go figure it out yourself how to do it" attitude. 130
hoblin @hobl.in · 18/07/2025We're very far from LLM's doing production-grade software development. Unfortunately. I would love to let it finish all the pet projects I've started and didn't have the time to finish. However, it can't even complete a single assignment to the quality I would accept. 110
