Module 13 Knowledge Check
5 questions. Pass mark 4 out of 5. Answer every question before checking the answer key below, then retry after reading the feedback.
1. The service should depend on…
- A. the vendor's SDK
- B. an interface the vendor implements
- C. the HTTP client
- D. the model name
2. A real model client with no configuration should…
- A. fall back to a fake
- B. fail loudly
- C. return an empty result
- D. retry
3. Model output should be…
- A. returned directly
- B. validated against a schema first
- C. logged in full
- D. cached
4. Cost and version information belongs…
- A. at the top level of the response
- B. in a separable meta block
- C. in the headers only
- D. in the logs only
5. The budget check should run…
- A. after the call, using the actual cost
- B. before the call, using an estimate
- C. daily, in a batch job
- D. in the client library
---
Answer Key and Explanations
Check these only after attempting every question.
1. B - an interface the vendor implements
That is what makes offline testing and provider changes cheap.
2. B - fail loudly
A silent fallback means a misconfigured production deployment serves fake predictions.
3. B - validated against a schema first
A confidence of 1.4 otherwise passes straight through and breaks your contract.
4. B - in a separable meta block
Top-level fields become part of the primary contract and cannot be removed later.
5. B - before the call, using an estimate
Checking afterwards means you have already spent it.
Practical Check
Apply this module to your own work: complete the module activity for *AI-Ready API Endpoints with Mock Model Services*, then write one sentence naming what your result shows and one naming what it does not.
Strong Answer Pattern
A strong answer names the task, the evidence used, the check performed, and the remaining limitation. It avoids "proved", "guaranteed", or "always" unless the evidence genuinely supports it.