DevOps with AI Agents in Practice
summary
A live agent prototype inspects a Git repo, creates Dockerfile, .dockerignore and docker-compose, and scaffolds a Python app plus PostgreSQL service, then runs containers and tests API/UI. The demo completes container build and basic end-to-end checks in about 8–15 minutes, with failures caused mainly by slow image downloads and flaky network. The system also reads runbooks, enforces action policies and secret scoping, records session decisions and can learn recurring fixes for future runs.
AI-written summary · about
chapters (7)
- 0:00Introduction and background
- 3:17Agent capabilities and setup
- 9:01Demo walkthrough start
- 12:38Docker environment setup
- 16:31Demo execution and results
- 20:10Testing and validation
- 22:13Conclusion and Q&A
related talks
31mAgents at Work: How GitLab Is Redefining AI Adoption for Engineering Teams
57mDeveloping Production-Ready Apps in Collaboration with AI Agents
33mThe Autonomous Enterprise
41mGoing beyond copilot with AI Agents
39mSimplify, speed and improve development with DevOps
34mThe Hard Truth About AI Agents in Production
41mGetting real DevOps ROI with your Generative AI
26mAI Agents as Your Co-Developer