Why should the origin not be publicly writable? How would you diagnose stale content?
Observable application lab
Deploy a minimal app, restrict access, and collect logs and health signals.
Evidence
Network rules, least-privilege role, CloudWatch observations, one controlled failure, and a recovery record.
Practice explaining
How do you distinguish DNS, network, application, and permission failures?
Repeatable recovery
Recreate a small lab with infrastructure as code and exercise a restore procedure.
Evidence
Versioned configuration, test output, RTO/RPO assumptions, secrets handling, and resource teardown.
Practice explaining
What did you restore and actually measure? What remains untested?
Label the scope honestly
Study
You can explain the concept. You have not necessarily deployed it.
Guided lab
You followed instructions. Record which steps were provided.
Independent project
You made and tested the decisions. Explain your contribution and limitations.
Production responsibility
Claim this only when you actually operated a real workload with permission and can describe the scope.
Sources and evidence boundaries
Program-specific details are transcribed from a user-supplied Batch 3 image reviewed on 23 October 2026. No official current intake announcement or application link was independently verified. This site is not affiliated with or endorsed by the program organizers.