Your developers already use AI. Your process does not know it.
Some of your team use an assistant every day and some not at all. Nobody can say how much of last month's code an agent wrote, what it cost, or which rules it followed.
This service is for engineering leaders who want AI in delivery as a deliberate practice, with rules and numbers.
What we set up
The kit in your repositories
The open-source kit installed and adapted to your layout, without overwriting what you have.
Seats and playbooks
Roles mapped to your people, each with a playbook that says what an agent may do in that seat.
Gates in your CI
Your rules turned into checks that run before a commit and before a merge.
A knowledge layer
Your documents and code indexed, so agents answer from your sources.
Measurement
A dashboard of use, attribution and cost, so you can decide from figures.
What you get
- A working setup in one pilot repository, then a rollout plan for the rest.
- A written brief and working agreement your team owns and can change.
- Training for each seat on its playbook.
- A first measurement report, and a review after the first weeks of use.
How we work
The kit is open source under the MIT licence, so nothing we install locks you in. You pay for the adaptation and the know-how.
- AI-SDLC Bootstrap Kit
- Claude Code
- GitHub Actions
- MCP