We're building a self-driving codebase in public, with daily livestreams until 25th of April.
High-performing engineering organisations are running multiple agents in parallel, in the background. But the teams pushing the frontier are going further: figuring out how agents can work together as a 'software factory'.
Since launching Ona Automations and the background agents landscape, the question we hear most is: how do you set a software factory up? Where are the limits? And what tooling and processes need to change?
This project is an attempt to answer that, in public.
Ona is a platform for background agents. Cloud dev environments give agents a place to run without depending on a local machine. Automations trigger agents on schedules, on PR events, so work continues around the clock. When an agent does something unexpected, we update the harness rather than fixing the code manually so that the repository and harness get smarter over time.


Schedule, recordings, and daily metrics (PRs merged, lines of code, test coverage, CI health) are all published to software-factory.dev. You can sign up for email updates, browse the source, file issues, or suggest tools and MCPs to integrate.
Find us on @ona_hq and @swfactory_dev.
Watch the kick-off call with Zach and Lou:
How 30 minutes of spec writing and 10 minutes of execution produced a PSX-styled 3D world with real Google city data.
How we replaced seven agent tools with one, moved from edge-triggered to level-triggered state, and built runtime guardrails to keep agents on track.
Time to first approval went from 2h 49m to 3.8 minutes. We let AI approve the PRs that didn't need human eyes.
This website uses cookies to enhance the user experience. Read our cookie policy for more info.