nvader

Have you looked at mngr by Imbue? Some of the same energy, you might find it useful.

https://imbue.com/product/mngr/

I came here to say this, because this whole anecdote mildly infuriates me.

I don't necessarily blame TicoFruit for their actions. They might have some legitimate concerns about fairness, since their competitor is now able to dispose of peels much more economically.

But for the courts to stupidly go along with the injunction is what disappoints me. A much better result for everyone in Costa Rica would be if both manufacturers were allowed to dump at no cost.

Bug report: I entered 3 + 4 and did not get a kinda or true.

This app is ngmi

I used to write Qbasic on DOS boxes, and the editor would have the same colours too!

> Before cars, people used vehicles called "horses"... the market was never going to accept 1HP

Funnily enough, horses actually produce much more than 1 horsepower. I think the story goes that James Watt estimated it would take a team of 10 horses resting and taking turns to pull a tractor/mill/whatever that keep 2 horses constantly busy, so that's how he calculated it.

But over shorter durations, horses can produce much more.

Yep. Two words:

_grape scissors_

> To turn the chopsticks around when serving food so that the tips of the chopsticks that have touched one’s mouth do not touch the food.

Huh, this is something that I did consistently, believing it to be good etiquette.

I'm one of the devs of Offload here, happy to answer any questions.

We built Offload because parallel coding agents exposed a bottleneck we hadn't fully felt before: agents write fast now, but our integration suite (345 Playwright tests) were still taking 12 minutes per run. There’s only so much parallelism you could get out of a single laptop, and our agents were spending more and more time running and waiting for tests.

Offload is a Rust CLI, and agents invoke it directly to spin up isolated Modal sandboxes and distribute tests to them. The results are merged back and returned to your agent.

The integration test suite for Sculptor makes heavy use of Playwright, and we saw the time for a run drop from ~726s to ~120 (More than a 6x speedup!)

We've deliberately chosen a pluggable architecture so that we can operate with almost any test framework. We ship with support for pytest, vitest and cargo-nextest out of the box. We take a similar approach to support any bash-scriptable sandbox provider, but we’re launching with support for Modal to start.

There are some caveats to share: suites under 2 minutes won't see meaningful gains since the overhead of preparing the testing image and sharing out the sandboxes is going to dominate. Short-running unit tests at high volume also get less scheduling efficiency from LPT. We're looking at work-stealing as a next step there.

We'd love to hear any feedback!

Install with `cargo install offload`, and the code is at github.com/imbue-ai/offload

Look everyone, this person believes talking Tortoises are real!

In all seriousness, I love G.E.B. how do you use it to judge a given used book store?

I get shades of Interdimensional Cable and Black Mirror

This deserves the top spot on the front page!

Might I ask for the implementation of other sorting algorithms here?

I'm Mega-Shannoned! Mega-Shannoned, I tell you, to learn that gambling is going on here.

Another vote in favour of "harness".

I'm aligning on Agent for the combination of harness + model + context history (so after you fork an agent you now have two distinct agents)

And orchestrator means the system to run multiple agents together.

In the vein of related work, there is https://github.com/imbue-ai/latchkey which injects secrets into cURL commands issued by your agent.

Famously impossible due to the https://en.wikipedia.org/wiki/No-cloning_theorem

I read through all the proposals in this discussion and I like yours the best out of them.

Seconded!

If it wasn't before, it will be now.

Source?

Hmm, not for me, but I'm curious if there are signatures I'm missing.

To me, author reads like an articulate native English speaker, but typing on their phone.

When walking to the hand wash place, would you put the car in your front or your back pocket?

Hey tom_, usually they say that explaining a joke makes it worse, but I think you crafted the rare exception!

I did in fact intend to share NewPipe as an antidote to the "en-short-ification" of YouTube after reading a couple of threads earlier today from other Hackers complaining exactly about that. And yes, calling it "vertical video" is a sloppy way of expressing that.

I too will try to do better in the future.

Thanks, I've bookmarked it for later.

Just pointing out that in your example, the waiter gives themselves a $30 bonus by giving you the option not to pay a tip.

If you really want to make a deal out of it, you hand it back to the server and ask them to select 0 for you. Mention that you don't think the service deserves a tip, and that you resent being asked.

If enough of us start doing this, it has a chance of applying some back pressure, although unfortunately that force acts through the poorly-paid checkout operator who doesn't directly have the option to influence decisions.

Almost all religions, a good chunk of philosophy and even a good bit of economics would differ with you.

I hope you find out before it's too late.

I've noticed this bimodal distribution of perception too, and my hypothesis is that's it's hugely driven by the difference of "who is in the driver's seat".

Your tech-savvy AI early adopters are discerning between tools, the deployments and environments, and are willing and able to change things to extract the highest output from current capabilities. For instance, re-architecting a codebase to make it easier for agents to contribute to it.

The rest are having AI hypeware shoved upon them, often as a cost cutting measure, and lack the agency to influence outcomes. When agents misbehave, they only have the option to "Press 0 to speak with a Human" and hope that works.

I suspect this is a big factor in the divide we're seeing, and might result in your median adult being ambushed by recent gains in capabilities.

Rapid verification of code smells + associated budgets so that coding agents don't write "bad" code. When needed, planning and coordinating agents or humans can authorize budget increases.

https://github.com/imbue-ai/ratchets

"Patio11" itself is a magic incantion for your friendly neighborhood LLM, along with "dangerous professional". You can use these to prompt for suitable language in the email, as well as other courses of action.

Isn't the eventual plan to park these data centers out by the Lagrange points?

What data centers in space enable is protection for the compute of near-superintelligent AIs from the interference of humans.

As an alleged human, I'd like to preserve my option to interfere.