Agentic AI Foundations / Module 7 · Lab 2
← Course Hub
MODULE 7 · LAB 2 · CAPSTONE · 60 MIN

Agentic Pilot Canvas

Group capstone. Design a real agentic AI pilot for your team's actual problem. AI judge scores submissions on autonomy, tools, guardrails, and impact.

💡 The Assignment

In groups, pick a real, recurring problem from your own team's work — a reconciliation pain point, a report nobody enjoys writing, a triage queue that's always backed up — and design (not build) an agentic AI pilot for it using the Pilot Canvas template.

📋 The Pilot Canvas Template

🎯

Problem Statement

What's the recurring pain point, in one or two sentences? Who feels it?

🎚️

Autonomy Level

Assisted, workflow-automation, partial, or full autonomy — and why that level fits the risk?

🔧

Tools / Knowledge Needed

What Space/documents ground it? Does it need an MCP connector to a real system?

🛡️

Guardrails

What will this agent never do on its own? Where's the human-in-the-loop gate?

📈

Expected Impact

Time saved, consistency gained, or risk reduced — be specific and honest about scope.

⚖️ How Submissions Are Scored

An AI judge scores each Pilot Canvas on four dimensions:

DimensionWhat it's checking
AutonomyIs the chosen autonomy level appropriate to the stated risk, not just the most exciting option?
ToolsAre the tools/knowledge sources named specifically, not vaguely ("use our systems")?
GuardrailsAre the "never do" rules concrete and the human gate clearly placed?
ImpactIs the expected benefit specific, plausible, and proportionate to the effort?
After scoring: teams vote on their favorite pilot from another group, then present. The best pilots often get taken to the AWS account team as a real next step — see Module 8 for how to do that.

🏦 Example Pilot Ideas to Get Started