Agentic AI Over Custom API Data | Dataford Interview Questions

Design an agentic AI system with a custom API data endpoint and custom database implementation to process data and present the user a query interface to ask questions about the processed data from the API data source.

Implement the deterministic query core as query_api_data(records, queries). Each record is a dictionary with an integer id; queries use field=value, field~term, AND, OR, parentheses, optional SORT field ASC|DESC, and optional LIMIT n. Return matching record IDs for each query, using indexes rather than scanning every record for each request.

Constraints

1 <= records.length <= 5000
1 <= queries.length <= 1000
Every record contains a unique integer id
Record values are strings, integers, or lists of strings
Query parentheses are balanced and operators are uppercase or lowercase
A query contains at least one field=value or field~term expression

Function Signature

def query_api_data(records, queries):

Your question is Agentic AI Over Custom API Data. 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

Constraints

1 <= records.length <= 5000
1 <= queries.length <= 1000
Every record contains a unique integer id
Record values are strings, integers, or lists of strings
Query parentheses are balanced and operators are uppercase or lowercase
A query contains at least one field=value or field~term expression

Function Signature

def query_api_data(records, queries):

Your question is Agentic AI Over Custom API Data. 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.