Agent swarms and the new model economics • Hacker News Board

Re: Agent swarms and the new model economics

#31

Post by kgeist » Mon, Jul 20, 2026, 9:15 PM UTC

Romario77 wrote: post #10

handfuloflight wrote: post #3

> To test that progress, we returned to a task the old swarm had struggled with: building SQLite from scratch, in Rust, from nothing but its documentation. Isn't SQLite's source code in its training data?

not in Rust. might be different enough ...

Even if no Rust code for it was seen during training, an LLM can trivially transpile SQLite's C codebase to Rust on the fly. For example, I just asked ChatGPT to write John Carmack's famous Fast Inverse Square Root algorithm in Erlang, without searching online or thinking, and it transpiled it immediately (while also extracting the knowledge in the same step). SQLite's semantics/code are stored in the middle layers of an LLM, and the last layers are able to convert it into any representation, as conditioned by the prompt. Cursor's experiment is deeply flawed because they merely extracted the model's compressed, lossy knowledge of SQLite's codebase and then just ran a bunch of tests/fixing rounds to make up for the lossiness. The claim that the agents built it from scratch is false.

Re: Agent swarms and the new model economics

#32

Post by pianopatrick » Mon, Jul 20, 2026, 9:18 PM UTC

I wonder what would happen if the agents did not have the spec. Like if you just said "build a simple SQL database that works as a single file", what would the agent swarm come up with.

Re: Agent swarms and the new model economics

#33

Post by Elad-Rez » Mon, Jul 20, 2026, 9:23 PM UTC

[flagged]

Re: Agent swarms and the new model economics

#34

Post by viccis » Mon, Jul 20, 2026, 9:27 PM UTC

mccoyb wrote: post #5

I find these blog posts (and the originals, with Anthropic's C compiler and Cursor's browser) somewhat funny, as if they have this enormous power to build ... but they can't build something unique or new. Like the software sucks, but look how powerful the process is (the models are indeed powerful). And it's a bit of a shame: by virtue of their position (their embedding in the fabric of venture capitalism), it seems…

The bottleneck has always been on the product side for 99% of cases. For the 1% where it's engineering, LLMs don't help as much.

Re: Agent swarms and the new model economics

#35

Post by bofadeez » Mon, Jul 20, 2026, 9:29 PM UTC

The issue is that the only models that could be trusted to work autonomously cost more than a human employee

Re: Agent swarms and the new model economics

#36

Post by adamtaylor_13 » Mon, Jul 20, 2026, 9:34 PM UTC

shay_ker wrote: post #24

How do we know if these models weren’t trained on Turso’s rewrite of SQLite in Rust? It seems both likely that they were and impossible to remove that code from pretraining. Doesn’t that make this just about LLM memorization of the training set? What am I missing?

Wouldn't that be in the authors' interest to disclose, if true?

Re: Agent swarms and the new model economics

#37

Post by ihsw » Mon, Jul 20, 2026, 9:48 PM UTC

Wowfunhappy wrote: post #23

svachalek wrote: post #22

Earlier quoted context omitted.

Some of this stuff is ludicrous. I finally tried /loop last week and discovered every loop iteration passes the entire context history to the model. So pretty quickly you're running a full 1M context window, without cache, likely just to check if something is ready or needs to be done. It's miserably terrible engineering unless your entire and only goal is to burn tokens.

Wait, why isn't context cached with /loop?

[dead]

Re: Agent swarms and the new model economics

#38

Post by voidhorse » Mon, Jul 20, 2026, 9:51 PM UTC

whatever1 wrote: post #30

jgalt212 wrote: post #11

Earlier quoted context omitted.

It's hard not read such quotes and immediately think of the Infinite Monkey Theorem. https://en.wikipedia.org/wiki/Infinite_monkey_theorem

If at the end Shakespeare-level literature is produced, does it matter whether we arrived there by random keystrokes?

Yes because the keystrokes aren't free and the world is finite. We need to pay these monkeys in bananas.

Re: Agent swarms and the new model economics

#39

Post by vessenes » Mon, Jul 20, 2026, 9:56 PM UTC

This is super fascinating, and I loved seeing testing of where exactly you need frontier intelligence -- looks like coordination / planning, but not coding right now -- the article's a bit of a tease, as we can't play with such a harness, or their new version control system or get a workable artifact out of it.

That said, I love the work on figuring out these harness coordination jobs. While there are analogs to human management there is also this enticing feeling that, since the models are broadly deterministic, we might be able to get repeatable science-type lessons about managing them with enough testing.

Re: Agent swarms and the new model economics

#40

Post by pzh » Mon, Jul 20, 2026, 10:02 PM UTC

Wowfunhappy wrote: post #23

svachalek wrote: post #22

Earlier quoted context omitted.

Some of this stuff is ludicrous. I finally tried /loop last week and discovered every loop iteration passes the entire context history to the model. So pretty quickly you're running a full 1M context window, without cache, likely just to check if something is ready or needs to be done. It's miserably terrible engineering unless your entire and only goal is to burn tokens.

Wait, why isn't context cached with /loop?

Because it's typically cached for 5min (1hr is a setting you have to explicitly opt into), and very few people run loops at a cadence of < 5 mins.