Validate LLM JSON Output | Dataford Interview Questions

Write a function to validate the structure of JSON outputs from an LLM response.

Implement validate_json_structure(response, schema), where response is a JSON string and schema maps required field names to type names: string, number, boolean, object, array, or null. Return True when the response is valid JSON, has an object at its root, and every schema field exists with the expected type. Extra response fields are allowed; malformed JSON, missing fields, and type mismatches return False.

Constraints

The response contains at most 10^4 characters.
The root JSON value must be an object.
The schema contains at most 100 required fields.
Schema type names are limited to string, number, boolean, object, array, and null.
Extra fields in the response are allowed.

Function Signature

def validate_json_structure(response, schema):

Your question is Validate LLM JSON Output. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

Write a function to validate the structure of JSON outputs from an LLM response.

Constraints

The response contains at most 10^4 characters.
The root JSON value must be an object.
The schema contains at most 100 required fields.
Schema type names are limited to string, number, boolean, object, array, and null.
Extra fields in the response are allowed.

Function Signature

def validate_json_structure(response, schema):

Your question is Validate LLM JSON Output. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.