Blog
Notes on multi-agent orchestration, prompt composition, and building with Weave.
We Merged 143 PRs in a Week. That Number Tells You Very Little.
One person and a set of coding agents merged 143 pull requests into Fleet in a single week. Here's what those PRs were made of, why PR and line counts measure effort rather than results, and what we'd measure instead.
Spend on the Environment, Not the Model
The biggest quality gains we've had from coding agents came from what surrounds the model: tests it can run, an app it can start and see, and checks that don't lie. How we set that up for Fleet, and where it failed us.
Understanding Tokens: How AI Models Count and Charge for Your Work
A practical guide to what tokens are, how providers bill for them, and concrete tactics to keep your agent costs under control.
Introducing Weave
A harness-agnostic multi-agent orchestration framework that brings structure, specialization, and safety to AI-assisted software development.