197 karma · joined April 20, 2022
Anecdotally we ran sonnet 4.6 for our more complex stuff and sonnet 5 was a LOT worse. 5.5 seems to have fixed it and we cut over our customer workloads. It’s strange, really.
54min in was what I was thinking of.
“Lots of people have tried, you’re welcome to try”
I apologize if I gave a bad gloss, but I think it was roughly in the spirit of what you said.
I think it does but there are weird dynamics I don’t fully understand. I’m curious about HNs thoughts.
My theories: Centralization around key projects due to AI pointing new users towards them. (At the same time this drives up the PR deluge onto these projects. Especially from newer users already heavily using llms.)
So many low effort AI-generated open source libraries that it becomes harder to tell signal from slop. More movement to the bigger projects because they are perceived as safer bets.
I don’t disagree with the premise, and I appreciate the roasts of the SV pseudophilosophers (he left out Ayn Rand tho).
Or, Perhaps the invention of the rocket emoji most likely was the cause of this phenomenon.
I’m not a booster a doomer or a boomer but I think it’s a reasonable litmus test for LLM coding to implement 80% of an existing app or service. It’s not an accusation against anyone using LLM (I do) nor is it an excuse to take shots, it’s just a way of framing SotA capabilities.
Weird article, perhaps I missed something.
To trust something like data-storage to vibe-coded nonsense is incredibly irresponsible. To promote it is even moreso. I'm just surprised you can't tell, too.