It's a coordination problem. 70 amazingly responsible adults still need to coordinate amongst one another. Ad hoc coordination and communication always breaks down eventually.
Well yes, ideally. But real world codebases aren't clean enough to be used as the example ideal. Styles change over time, there are always code migrations and refactors in flight, legacy code exists, etc. Using specific examples of what you expect the LLM (and humans) to do now is necessary.
This is fine for one off tools and I do the same. But building long-lived "professional grade" production software this fails real quickly.
My team is using AI for most of the code, but the human review layer is crucial and unavoidable if you're interested in things like reliability, uptime, controlled feature rollouts, the integrity if your user's data, etc.
Yea. Modern Ruby is "fast enough", but it's very real that when Ruby was hitting its peak it was dog slow. It's hard to shake those sorts of reputations (similar to the "can't scale" reputation that Rails got because of Twitter)
I feel confident in saying that I am better at computers than 99.99% of the general population and I have no clue what “SS” or blue USB ports are supposed to indicate.
This was such a weird mention to see in the article. Stripe Atlas is a service that helps new businesses incorporate and onboard onto Stripe/partner services with some startup credits. It's been around forever, has nothing to do with AI, and is generally a very well-respected service.
If they’re reaching the same results across a variety of the most popular public models, it doesn’t seem like that big a deal to know if it was Opus 4 or Opus 4.5
Yea this feels like saying “if you give them good enough specs they’ll produce the code you want” which reduces to…writing the code yourself. Just with more steps.
I think they're just saying that data extraction tasks are easy to evaluate because for a given input text/file you can specify the exact structured output you expect from it.
I think your post is fair but it's worth pointing out that learning via watching is much less effective than learning via doing.