About us

We’re a small team on a mission to make software development more efficient and more enjoyable. We come from agencies, from large enterprises, and from startups. The day-to-day challenges in each were different, but the underlying problems were usually the same.

We started by asking how AI and agentic coding agents would change the way we write software. As engineers, we could feel the shift already happening, and we wanted to understand what it meant for us.

Our first direction was context engineering. It didn’t take long to realise that wasn’t where we could add real value. But while we experimented, we noticed something else: a problem that had bothered us for years might finally be solvable. How do you keep a project’s documentation current and relevant?

Software developers have a bad reputation for documenting their work. Most projects have no real record of the decisions and architecture that shaped them, and what does exist is usually several versions out of date. This isn’t laziness. It’s time constraints and delivery pressure taking their toll.

Infinite documentation isn’t the answer either. If your docs run as long as War and Peace, no one is ever going to read them. You need the right information at the right time, in context. Tools for storing and querying documentation have been around for years, but most of the attention has gone to presentation. Agentic workflows change that. We no longer set out to describe a whole product at once; we work on much smaller pieces at a time. So storage and querying grow in importance, while the presentation layer shifts underneath them: when search gives way to chat, the way we consume, store, and argue about knowledge shifts with it.

That is the problem we’re working on. How do we make information make sense to humans in this era, and to the agentic counterparts they work alongside.