Detail

Detail is building an AI engineer to solve technical debt.

We got into software because we liked making things, and we found the reality of professional software engineering to be disappointing:

• spending weeks on simple features because we needed to cut across legacy code
• having meetings to find out if a change is safe
• multi-quarter projects to unwind a technical decision someone made years ago

Every engineer should have the joy of a productive codebase, and it shouldn't take a rewrite to get there.

Our approach uses a new kind of code sandbox to simulate prod. We backtest changes by re-running prod traffic in isolation. That means we can author pull requests that are reliably correct and ready-to-merge. We can solve all sorts of codebase problems this way.

If you're interested in joining, please get in touch: [email protected]

We're especially keen to meet you if:

• you've tried out lots of dev tools and have opinions about them
• you're excited about the prospect of going deep on language runtimes
• you've been frustrated by the difficulty of shipping changes that should be simple

We are grateful for support from Conviction, OpenAI, Guillermo Rauch, Olivier Pomel, Dev Ittycheria, Arash Ferdowsi, Alex Graveley, and many other outstanding people.