Would a human end a long-winded text with…?

First I thought CC wrote all its code, but it’s about the engineer’s contributions to CC, which is quite different.

I am curious, what’s the point of re-running these interactions on a UI?

Reproduction I suppose. I would like the same things as OP too.

LLM outputs are qualitative; they can't really be automatically scored and prompt enhancements tend to multiply the bug. It can solve a problem, but introduce a new one. It's practical just to do it manually.

Kubernetes.

In this argument, you are selling OpenAI a bit short.

Public speaking is great, but not sure if it’s easier to keep a recorded talk (or even a company blog post) up to date than anything else you have full control over.

I think the "keeping up to date" is a fools errand IMO. Because you end up with "half-up-to-date" documentation where someone thought to update some part of it, but not another. And it gets incoherent.

So my preference is a coherent story at a point in time

How about Asahi Linux, or a Fusion/Parallels VM on macOS?

I am waiting for someone with an AI tool that visualises the actual trend.

Have you even tried searching for the term?

Anyway, here you go: https://modelcontextprotocol.info/specification/