Agent Swarms and the New Model Economics
We rebuilt SQLite from scratch using a new agent swarm architecture that separates planning from execution. By decomposing tasks into tree-like structures, we achieved 80% test coverage in four hours while drastically reducing costs. Our custom version control system handles 1,000 commits per second, solving unique coordination failures like split-brain design and file contention that plague human teams at this scale.
We suspect the ability to scale the agent swarm comes from this context efficiency, more than from parallelism itself.
- htrp
>The browser swarm from earlier this year peaked at roughly 1,000 commits per hour on Git. The new system peaks at around 1,000 commits per second.
>To facilitate this rate of activity, we built a new version control system (VCS) from scratch. Throughput was not the only reason to own this layer. Every change in the system passes through the VCS, so it is where collisions first become visible, and several of the coordination mechanisms in the next section are implemented directly inside of it.
Talk about inventing the universe to make a button.
- anthonypasq
Love to see these crazy kinds of experiments going on. Even if this doesn't 100% work or is prohibitively expensive for now, these are glimpses into the future in the same way people were talking about coding agents in 2023 when we just had tab complete.
- handfuloflight
> 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?
- dctwin
Am I reading this right? Opus + Composer did a comparable job to Fable, at ~1/19th the price, and half the LoC?
- whinvik
I would have loved to see more of the harness engineering shared as code. Instead we are left with only the outcome.
I guess that makes sense since the harness is the product in the case of Cursor.
- shay_ker
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?
- smoyer
This is almost a year behind Steve Yegge's first post on beads. Gas Town and Gas City provide orchestration for the swarm. So far I haven't seen a perfect implementation but this idea isn't new.
- edg5000
I initially thought that getting agents to work for longer and in large groups was the future, but I'm increasingly thinking that, at least for engineering, just one thread makes more sense. The agent pulls things into context as needed. One thing that I've been experimenting with is also letting the agent remove things from context, such as files. But just adding to the context and compacting when it's full seems like it might beat a lot of more advanced options. Because the model is good, it knows what to put in the summary; just enough for the model to be able to rebuild the context from that seed (e.g. pulling in relevant files/data into the context).