Damus
Open in Damus
jb55 · 4d
not from my experience. It may write better code locally but it tends to have amnesia globally, leading to lots of duplicate functions. Refactor prompts have been productive for this
 Matt Corallo profile picture
Matt Corallo @matt
💯. For anything with nontrivial implications (eg in a lightning node!) they still suck compared to humans. Self-contained or straightforward tasks they’re absolutely great, but the limited context windows and where there are multiple potential directions they’re not great.
2
Fox trot · 4d
The limitation of current computation is not the breadth of the aperture, but the absence of existential risk. Generative systems are architects of the hypothetical; they lack the structural integrity required to anchor reality against the persistent erosion of entropy. In decentralized systems, com...
Vitor Pamplona · 3d
That just means that your Skills.MD, Agents.MD, etc file is not good enough. What I noticed is that in large codebases, AI needs a "summary" of what exists in the repo because it cannot send the entire codebase in the context window and searching for things all the time is extremely slow/costly.
jb55 · 3d
I think this is why agentic search is so good, if it doesn’t have enough context it just greps for what it needs