←
Yaron (Ron) Minsky8,665 posts
Yaron (Ron) Minsky avatar

Yaron (Ron) Minsky

@yminsky

Occasional OCaml programmer. Host of @signalsthreads. @yminsky@discuss.systems @yminsky.bsky.social threads.net/@yaronminsky

◷ Joined June 2009▧ 210 Photos and videos
Laurence Tratt avatar
⟳ Yaron (Ron) Minsky retweeted
Laurence Tratt@laurencetratt·Nov 17

I wonder if David is seeing the same thing I am: LLMs of today are tactical geniuses but still learners at strategy. A classic problem is putting fixes at the wrong level in the system: a short-term tactical win but a strategic loss. Those strategic issues compound in weird ways.

marius eriksen avatar
⟳ Yaron (Ron) Minsky retweeted
marius eriksen@marius·Nov 17

I'm struggling with this too. Most code is fundamentally utilitarian, and agents getting it done is pure good. But agents will build on whatever foundations you give them -- and I think those foundations matter more than ever.

David R. MacIver avatar
⟳ Yaron (Ron) Minsky retweeted
David R. MacIver@DRMacIver·Nov 17

After a ~year of heavy LLM use in library development I'm currently trying to ship my first production system with 100% LLM-written code, and... I'm definitely finding that they're bad at this in ways that correspond pretty well to being bad at software design.

Yaron (Ron) Minsky avatar
Yaron (Ron) Minsky@yminsky·Nov 16

Either way, I don't think it's just small mindedness. The inability of current agents to generate clean systems with beautiful abstractions is a real cognitive gap! Maybe it will be closed soon, and maybe it doesn't matter, but the gap is real nonetheless.

Yaron (Ron) Minsky avatar
Yaron (Ron) Minsky@yminsky·Nov 16

But...can they? We don't really know how effectively they can operate in the face of what we think of as bad design. And we don't have large scale, complex projects that have been entirely vibe coded and have shown a lot of impact and success.

Yaron (Ron) Minsky avatar
Yaron (Ron) Minsky@yminsky·Nov 16

One perspective is: it's small minded of us to care too much about traditional notions of taste. We should just care about the results, and if agents can operate effectively without clean abstractions and well factor code, it's just prejudice to care.

Yaron (Ron) Minsky avatar
Yaron (Ron) Minsky@yminsky·Nov 16

One interesting bit of AI discourse: it's pretty broadly agreed that agents aren't great at software design. But...they're still good at getting tasks done. So, maybe software design doesn't matter? I think there's a lot of complexity baked into this question!

Annemarie Gray avatar
⟳ Yaron (Ron) Minsky retweeted
Annemarie Gray@annemariegray·Nov 15

+1 for having a kid in NYC: when your baby wakes up at 6 am on a Sunday you can take them to the marathon reading of The Power Broker

Post attachment
difficultyang avatar
⟳ Yaron (Ron) Minsky retweeted
difficultyang@difficultyang·Nov 14

I just want to state what it means to be deadlock free HOW HARD COULD IT BE

Yaron (Ron) Minsky avatar
Yaron (Ron) Minsky@yminsky·Nov 13

A new Signals & Threads! Come for the discussion of integrating LLMs into chat trading applications, stay for the harrowing story of Chloe's participation in a harrowing 10-day horse race in Mongolia.

Post attachment