jwolfe

That's versus Opus 5.5 max effort, no? Opus high has the closest cost and intelligence characteristics and generates far fewer output tokens per task than the one Kimi 3.1 Pro effort level we have data for.

This is very cool. Unfortunate that it gives the wrong pronunciation when words get split in a line wrap (e.g., 大学 got split between lines and changed to incorrect furigana for me).

The Internet says they have a 27% stake. Where'd you get your number from?

A whole section of the article is about this. "Can Land Value Tax Be Passed on to Tenants?"

This definition of cost is not particularly useful. You want cost per task, not cost per 1m tokens. Artificial Analysis does a good job of this.

Yes, the first sentence of the article.

This is true, and it's a good point. I agree that open weight model regulation would either limit the intelligence of open weight models below the frontier or kill them entirely. It would not prevent closed weight Chinese models, but those aren't really tenable in American enterprises unless they strike deals with American cloud providers to deploy them, via products like Bedrock. I unfortunately also have seen no evidence that we can put any sort of guardrails on open weight models whatsoever and so am reluctantly convinced that they should be regulated below the frontier until such a time as someone comes up with a mechanism that is not circumventable.

Please elaborate on what the AI labs are specifically requesting and how that results in slowing down Chinese model progress below the frontier.

That's what pacing the frontier is, and is what the labs are pushing for.

Please explain how putting an upper bound on how good the strongest models can be prevents cheaper less strong models from catching up, rather than enabling it. I do not understand this argument at all.

claude.md predates agents.md.

Every turn means more tokens in ways that are not obvious to most people and lead to tons of unnecessary cache reads.

No harness can batch your agents.md read with the reads the contents of the file tell it to read.

This wastes both tokens and turns. But yes it's probably the best option we have today.

3 of those are already in there.

I can't imagine that they had managed access to the internet but could not figure out how to contact anyone at the company if they wanted to.

That is apparently the existing academic term. It's got a Wikipedia page. But yes, all buzzwords were invented.

If bash placed the current directory in your PATH by default, then yes.

You're in luck: https://github.com/apple/foundationdb/blob/main/design/corou...

They're referring to Enterprise customers, though should have been clear about it. Enterprise plans on Claude for example no longer include any baseline tokens. It's 100% usage based pricing.

That's not a very good tldr. The answer claimed in the paper is that the combination of the two is better than either alone.

Fortunately the economics of infinitely copyable software aren't going anywhere. If there is actually a sizable market, it can be served with relatively little cost, compared to hand-producing physical goods. I am not fully convinced there will be such a market, though.

Your work laptop documents are not private personal notes.

This is also partially why I do not log in to work accounts on personal devices or personal accounts on work devices.

You commonly receive very close proxies for diagnoses through MyChart already when results come back from the lab.

> If it's something like "Refactored the apartment list service improving P99 Latency from 2s to 180ms", it definitely boosts the resumé in my mind. A good engineer would be measuring their impact and likely have numbers like that off the top of their head.

> But if it's like "Increased revenue by $18.7M by reducing time-to-first-interaction latency from 2.3s to 117ms, increasing conversion by 47% and LTV by 28%," with the same fidelity on each bullet, I'm very skeptical.

Do you mind explaining why? The former doesn't indicate caring about business impact whatsoever (is this service in the critical path of any online process? Who knows!) while the latter does.

The comments on the article include other people replicating all or parts of the finding. I'm also pretty confident Kelsey Piper wouldn't fail to disable memory while simultaneously talking about how Claude incognito mode is insufficient to prevent the app from handing it your name.

Pre commit hooks exist. People just don't like being prevented from committing for reasons such as this.

This is a user preferences setting for what it's worth.

In general these agents support LSPs, which is often as much information as your IDE will give you. They are also not required to output syntactically correct code token by token when running agentically, because the loop is:

1. code

2. syntax check / build / format / lint (details language dependent)

3. test

and they can hop between 1 and 2 however many times they want.

> I don't think it's a psychologically positive self identification to see yourself merely as a gatekeeper and toll extractor rent seeker who only makes a living by withholding agency and skill from others.

That's an incredibly uncharitable reading of the parent comment. At no point in history prior to maybe this year could you argue that working in software was gatekeeping, toll extracting, or rent seeking. Being a highly skilled craftsperson creating software for those who can't or don't want to is a very psychologically positive self identification. Lamenting that the industry is moving away from highly skilled craftspeople is also perfectly valid, even if you believe that it is somehow good for society, which is yet to become clear.

This thread is about bringing these people to the US.