> 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.

> But over shorter durations, horses can produce much more.

A human can generate more than a horsepower, too, for short times. I used to have how fast you'd have to go up a flight of stairs to have done so.

You can redline a car, too, but that's not the engine's working horsepower.

However, Watt was a salesman, and intentionally misrepresented "mine pony power" as "horse power".

No but HP specs are always peak power in ICEs, and usually peak power in electric.

Some more honest ones state steady state in electric, but they are the exception.

Yep. Two words:

_grape scissors_

Look here, Bridgerton: https://www.youtube.com/watch?v=NNegQyn-4N8

> 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.

Somewhere on the page they mentioned that there are separate serving chopsticks. Turning the eating chopsticks around is probably more normal when there aren't separate ones.

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?

In my experience book stores either have multiple copies or zero copies.

Robertson Davies is another bellwether author I should mention.

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.

This has also been my understanding of all of these terms so far

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