Studio hour·45 minutes

Studio Hour: Model Interface Contract

Specify and test a small structured model response before adding retrieval or tools.

Scenario

A fluent assistant sometimes omits citations, invents status values, or returns prose where the application expects a typed object.

You will demonstrate

Define a bounded schema.
Reject malformed responses.
Specify fallback behavior.

Project evidence

Show the work, not a checked box.

Each response is stored in this browser as you type. Include metrics, test output, or a decision rationale wherever the deliverable asks for it.

Publish the interface contract

Define required fields, enums, null behavior, one valid example, three invalid examples, retry policy, and user-visible fallback.

Required evidence: A compact schema test table with deterministic pass/fail outcomes.

Your evidence

0/80 minimum characters

Runnable Python lab

Reject an invalid model response

Implement a deterministic validator for required fields and allowed status values.

Console

Run the code to inspect its output. Use Check solution only when you want to run the exercise checks.

Code runs only in a time-limited browser worker. It is not sent to the IntuitiveAI server. Do not paste secrets.

Expected result

True False
ALL TESTS PASSED

Project defense

What should the application do with a response that fails its schema?

Rubric self-review

Rate the evidence, not your effort: 0 missing, 1 weak, 2 adequate, 3 strong. All criteria must be reviewed, but a low honest score does not get hidden.

The contract makes allowed output, validation failure, retry limits, and fallback behavior explicit.

Useful references

JSON Schema specificationFree specification and learning material for validating structured data contracts.

Project completion gate

Completion is controlled by stored evidence, deterministic tests, a decision defense, rubric review, and the artifact when required.

Evidence pendingCode pendingDefense pendingRubric pending

Optional cloud portfolio

Submit evidence across devices.

An account is required. Submit only when the local completion gate passes. AI review is advisory and separate from deterministic completion.

After submission, I may choose to send the stored evidence text to the configured AI model for advisory rubric feedback. Artifact bytes are not sent by this review action.

Account settings

Back to Topic Library