Transition from traditional ops to agentic DevOps workflows: Review

The core 'exercise' here is a multiple-choice quiz. You read a wall of text describing a hypothetical scenario—migrating a legacy app—and then answer questions about how you'd instruct a team of AI agents to handle it. No code, no YAML, no actual interaction with an agent. It’s a PowerPoint deck pretending to be a lab.

I see what they’re trying to do. They're building a mental model for orchestrating agents before the tools are fully baked. The diagrams showing a 'Planner' agent delegating to 'Code,' 'Test,' and 'Deploy' agents are fine for a whiteboard session. The module introduces some useful vocabulary, like distinguishing between an agent's 'instructions' (the goal) and its 'skills' (the functions it can call). This is probably the one practical takeaway: it gives you a structured way to think about prompt engineering for these systems.

But that’s it. The whole thing feels like a high-level design doc for a product that doesn't exist yet, or is maybe in some private preview. I spent 90 minutes reading what amounts to a long-form marketing brief. It defines the problem space but gives you absolutely nothing to solve it with. If you're looking for an architectural overview to kill a lunch break, fine. If you came here to learn how to actually build or transition a workflow, this isn't it. You leave with concepts, not capabilities.