About the playbook
This playbook is about how software engineering and R&D teams are redesigning their practices to build and ship software with agentic AI. Every practice here is grounded in what engineering leaders here at GitLab and among our customer community are actually doing — not hype or demos that never scale past one team.
As agents take on more of the coding itself, engineering leaders and their teams have to redesign the practices around it: finding where delivery actually slows, deciding what people and agents each do best in the rest of the software lifecycle, setting the standards that let agents work safely in the codebase, verifying what agents produce before it ships, and growing the engineers who now direct them. Becoming AI-native is a practice that leaders keep refining with their teams as tools and understanding evolve.
Go deeper and you'll find this playbook organized around three parts of that practice:
- Redesigning the work so agents fit the flow of delivery instead of attaching to steps built for people alone
- Working with agents by deciding where human judgment and accountability stay and how engineers keep growing as the coding shifts
- Governing and scaling from one workflow to many with the context, verification, and measurement that let you move faster without losing control of quality or risk
The guidance is written to apply across teams and tech stacks while staying concrete enough to act on. These pages will change as the practice does, so treat them as a current, shared reference rather than a finished manual.
This playbook covers the practices. GitLab Well Architected covers how to set them up in GitLab.