This course is for DevOps practitioners who already know the fundamentals and want to see where Google Cloud’s AI tooling actually fits into a real pipeline, not a lab exercise, not a toy demo. You will not be training models or doing MLOps. Instead, you’ll treat Gemini and Vertex AI as tools inside your existing DevOps workflow, the same way you’d treat a linter, a monitoring dashboard, or a deployment gate.
The course follows one continuous demo app, a minimal Flask service called crash-app, from its first deploy on Cloud Run through a full AI-augmented Cloud Build pipeline. Along the way you’ll containerize and deploy on both Cloud Run and GKE, compare deployment strategies, provision infrastructure with Terraform, and then layer AI directly into the pipeline: a Gemini-powered code review gate that runs on every pull request, automated log summarization after deploy, and an incident response flow that turns a crash alert into a Pub/Sub-triggered AI triage message in your team chat.
By the end, you’ll have a working, end-to-end Cloud Build pipeline with seven real stages, deployed against a live GCP project, that you can adapt directly to your own team’s workflow.
