After two months of heavy "coding" with AI agents, I have one conclusion: if your codebase already exists, is fully human-written, and you use agents to add or improve features, it works great. However, when you try to create something new from scratch, they tend to add so much overcomplicated spaghetti code that it's hard to maintain in the long run. No matter which coding model you use, sooner or later, you'll hit a wall you can't break through.
Ok, I think my experiment leaving AI working on stuff 24/7 ends here. It doesn't work. Code explodes in complexity, results are not that great, the AI can't get past hard walls (it is still completely unable to even *grasp* SupGen), and it is insanely expensive (spent ~1k over the last 2 days). The best results are on the JS compiler, mostly because it is familiar (compared to inets), but not worth losing control over the codebase.
I think the dream of having AI's working on the background and making real progress on things that matter (i.e., truly new things) isn't here yet. It is still a machine hard-stuck on its own training data, incapable of thinking out of the box. It is great for building things that were already built. But not new things
Also coding normally has the under-appreciated advantage that you're doing two things at the same time: building a codebase *and* learning it. AI's do only half of that. The other half is obviously impossible 🤔
“AI is the future!”
Nah, AI is YOUR future because you’re a lazy, talentless wannabe. I still have a properly functioning brain so I’m gonna keep using that.
Our engineer flat-out refuses to use AI IDEs like Cursor or Claude Code. He will use LLMs to look things up or proofread, but he will not let them touch the actual code.
That is basically turtle speed compared to using an AI IDE, so yeah, we ship slower.
But at least I can sleep at night knowing the codebase will not slowly turn into auto-generated spaghetti that nobody really understands and that becomes expensive technical debt later.