Post #1688712

2026-04-01 11:37 UTC

@maxmortillaro @jonny True fact: I was recently asked why I am not using more AI tools. I am incredibly open about trying new tools and have indeed tried lots of AI tools with varying levels of failure. I was told the tools have gotten better and I should try again. I tried again, with varying levels of failure. I’m not sure that some of these people have actually used these tools to do things or if they are simply parroting what other people they respect have said.

Replies (1)

@jonny@neuromatch.social 2026-04-01 11:52

@sfoskett @maxmortillaro Ive tried them over and over again and try all the whacky rigamarole skills and mcps and everything that gets thrown against the pipe. They can be useful as expensive rubber ducks, limited restructures that fall just outside the tedium of regex and keyboard shortcuts, boilerplate, and other tasks that are in-domain for statistical models of token sequences. I have done several "me first, then LLM" comparisons where I write the prompt text I would use to describe the feature, write my best version first, and then without showing it to the model, give it my honest best attempt to use the pre-written prompt starter and keep coaching it without giving away my implementation. Never once has the LLM ever managed to write a coherent feature, no matter how small and well defined it is. Even when I cheat by reminding it of some caveat that I realized during my own writing, it cannot take into account the existing code and find the smart path from here to there because that's not what it does: it generates text in waves of context, and so all it can do is lay down a new wave. I need to write those up and do a few more systematically and observed.

Open ##1688713