Agents at Work: How GitLab Is Redefining AI Adoption for Engineering Teams
summary
Claims that agents execute event-driven workflows inside GitLab and actually ship code: the Duo Agent Platform reads an issue, runs an issue-to-MR flow, produces a merge request (demo shows two commits) and returns code-review feedback in minutes. It diagnoses failing pipelines and can create a proposed fix via the fixed CI/CD flow, integrates with VS Code/CLI, supports SaaS/self‑managed/private deployments, and requires human approval before merging. The demo project is a Python Flask Strava app and the generated changes run locally.
AI-written summary · about
chapters (14)
- 0:00GitLab’s AI adoption intro
- 1:55DevSecOps platform overview
- 4:02GitLab Duo Agent Platform
- 5:50AI workflows beyond code
- 7:58Agents and flows in Duo
- 9:42Duo Agent deployment options
- 11:48Duo Agent use cases
- 13:56Live demo: Duo Agent
- 20:18Automating merge requests
- 22:03AI code reviews & customization
- 25:09Fixing failed pipelines
- 27:13Custom agents in GitLab
- 28:38Autonomous AI workflows
- 29:41Next steps for GitLab AI
related talks
18mUsing AI Across the Dev Lifecycle
19mAgents at work: plan - test - CR - deploy - repeat
01mGovAI Summit Spotlight!
51mBeyond Code Generation: Governing AI Agents Across the DevSecOps Lifecycle
23mFuture of the future: creating an AI product beyond hype
24mEmpowering the GitOps Future with AI
24mDevOps with AI Agents in Practice
26mAI Agents as Your Co-Developer