popkkksidhd

A big reason LLMs can write code productively is cause software dev tends to be pretty resilient to failure. And at the pessimistic end, lots of companies can make money off buggy products, so it’s easy to hand wave the failures away. Does that hold true for architecture?

That being said, I’m certain they’ll be extremely useful in other areas such as review, auditing, etc - if those even matter for architects.

Even Isaac Newton fell victim to complex financial bubbles.

> I spend a ton of time planning out & specing the problem with claude, to be as specific as possible and be as constrained as possible so it cannot go on side-quests

I still have yet to see this be efficient. The time I spent doing this is way less efficient than simply giving the LLM some high level direction and putting that time towards reading the code it generates and correcting small piece by piece (occasionally doing things manually, but that’s only for more off the rails stuff).

It’s about the same amount of time as up front planning, except I know the code much better as well as end up with less mistakes compared to having it implement a front loaded plan.

Knowing the codebase makes everything else 10x more efficient too - triaging bug reports, planning other work, knowing what work needs done (this is a big one), reviewing PRs, etc.

And who caused the supply crunch? Very deceitful framing. Worse than stating plainly what he means like the headlines have done.

He could very easily solve the shortage by not buying up the entire supply. Further, the frontier labs could even coordinate in favor of consumers and agree to limit their compute purchases.

Is it really that grim? AI helps a lot and is great at certain well defined tasks, but the “AI team” at my small org launched a new product very fast and very buggy. Customers were pissed. They had to go back to reading the code and applying some thinking and care to get a working product out. Still faster don’t get me wrong but there’s still a lot of tradeoffs to be made that AI can’t really do alone. Like there’s no objective correct answer.

Is this just domain dependent?