Platform AI Assessment — Your Environment
Use this template during and after Module 2 to document what AWS platform AI features are available to you, what they can do, and — critically — where they fall short.
Services Evaluated
| AWS AI Feature | Available in Your Account? | Free Tier? | Capability | Gap (What It Can't Do) |
|---|---|---|---|---|
| CloudWatch Basic Metrics | Yes / No / N/A | Yes (always free) | Monitor resource utilization, set threshold alarms | Can't correlate with deployments, no runbook integration |
| CloudWatch Anomaly Detection | Yes / No / N/A | No ($0.30/alarm beyond 10 free) | Learns baselines, detects metric deviations automatically | Can't correlate with deployments or follow runbooks |
| Cost Explorer | Yes / No / N/A | Yes (web UI) | Cost trends, service breakdown, forecasting | Can't recommend specific optimizations for YOUR architecture |
| Q Developer | Yes / No / N/A | Yes (Builder ID, 50 req/month) | Code explanation, security suggestions, IaC review | Limited to code context — can't analyze live infrastructure state |
| DevOps Guru | Yes / No / N/A | 3-month free trial (expires) | Multi-service anomaly detection across CloudWatch/Config/CloudTrail | Can't execute remediation, limited to AWS services only |
Your Observations (Fill In)
CloudWatch
What I saw in the alarm data:
- Alarms in ALARM state: _____
- Metric that triggered it: _____
- Information I needed that wasn't there: _____
Cost Explorer
Top 3 cost drivers in my account (or mock data):
- _____ ($___/month)
- _____ ($___/month)
- _____ ($___/month)
The spike I can't explain without more context: _____
Context I'd need to give an AI agent to investigate this:
Q Developer
Task I gave it: _____
What it got right: _____
What it didn't know that I know: _____
Key Gaps Identified
List the 3 most significant gaps you found between what platform AI offers and what you actually need:
- _______________ (e.g., "CloudWatch can't correlate CPU spikes with our deployment pipeline")
-
-
What a Custom Agent Could Add
For each gap above, describe what a custom agent (with tools and domain context) could do:
- _______________ (e.g., "Agent reads alarm, checks git log for recent deployments, cross-references with runbook, creates Jira ticket with root cause hypothesis")
-
-
Your Platform AI Readiness Score
Rate your organization's current platform AI usage (1=not using it, 5=using all available features):
| Feature | Current Use (1-5) | Automation Potential (1-5) |
|---|---|---|
| CloudWatch monitoring | ||
| Cost visibility | ||
| Code assistance (Q/Copilot/etc.) |
Overall: The gaps you identified above are your custom agent opportunity list. You'll build a solution for one of them on Day 3.