These are some very extreme examples of this that push the feature's limits:

https://news.ycombinator.com/item?id=47256810

https://news.ycombinator.com/item?id=26495059

At Nvidia, we have no limit for Anthropic or Open AI models (for now) and are heavily encouraged to use them as much as possible.

The fact that they've started promoting using the Caveman mode tells me that the unlimited usage policy is taking its toll.

Fwiw, nobody has ever suggested to me that I employ token compression in my daily workflow. I don't pay full attention in all the AI workflow demos I'm supposed to attend, but I don't recall that even being discussed. Is this an Nvidia blog or tweet you're referencing? I'm actually interested to see what they have to say.

What is Caveman mode?

Please don’t tell me you’re writing RTL

I'm not, I work higher level products. I've talked to a few people who do but I don't recall if they have different standards.

I learned to read Devanagari by putting generic Google images cards on my phone's wall paper.

I had never heard of low background steel. That's a fascinating problem it solves (and I love the analogy).

In the bay area, I've met relatively few NRIs who don't know Hindi well, even if it's not their first language. Most of them that I've met are not even Kannadiga, Mallu, Telugu, or especially not Tamil. Sample size of at least several dozen.

Studying Hindi has felt very rewarding to me, and it impresses people disproportionately to my actual skill, but I don't feel it has affected my ability to communicate with coworkers whatsoever.

Any media company which deliberately rids itself of everyone willing to speak vaguely positively of transsexual people may not be attracting the most free thinking writers.

That feels reductive. It's actionable to get OCI, renounce US citizenship, and become Indian after a decade. It's just not very practical for most people. I know many people who don't work their dream job, but switching to a favorable company isn't currently practical for them.

My understanding from speaking with a few Tile IR devs on dates is that its primary motivation was providing better portability for programming tensor cores than PTX offers. Nobody ever told me they saw it as a response to anything other than customer feedback.

Every reasonable language has a Python interop story. All it takes is C FFI. But what Mojo promised early on was the eventuality of compiling a large amount of Python code if not entire wheels as Mojo.

I don't recall they promised that. They promised it'll be a superset, but Mojo introduces new keyword. Mojo could support all Python features today exactly as they're supported in Python and you wouldn't still be able to copy Python code into Mojo and compile it