What you will produce

A one-page deployment plan for a small application. Your goal is to make your decisions understandable before you create infrastructure.

1. Describe the workload

Choose a simple application you understand: a portfolio website, an internal dashboard, or a small API. Write down who will use it, what information it handles, and whether it needs persistent storage. Keep the first version narrow enough to explain in two minutes.

2. Draw the request path

Describe how a request reaches the application and how the application reaches its data. Identify which components need public access and which should remain private. Record the assumptions you will verify during implementation instead of treating them as facts.

3. Define access and recovery

List the human and application identities that need access. For each one, describe the task it must perform and the permission it needs. Record where configuration will live and how you will avoid committing credentials to source control.

Describe one failure you expect to recover from. What signal would reveal the problem? What action would restore the service? What data, if any, could be lost?

4. Set a spending boundary

Write down the resources you expect to create, the costs you need to check, and how you will remove the resources after the lab. Set a budget alert in the account you use. Do not assume that a small resource is free.

5. Review your plan

Ask another learner to explain the request path back to you. If a decision is unclear, improve the document before adding complexity.

Submit your evidence

Keep a README containing the architecture description, assumptions, access decisions, recovery plan and cleanup checklist. Include a short explanation of the tradeoff you found hardest.

This learning pilot is a practice exercise, not a production deployment instruction or a certification guarantee.