Build a support lookup before an AI agent
=========================================
Python developers choosing an agent course

Original practice exercise by AI Course Rank. No purchase required.

STARTING MATERIAL
Public sample catalog:
A1 | Trail Lamp | 29 USD | 14-day return | indoor use only
B2 | Travel Mug | 18 USD | 30-day return | hand-wash only
C3 | Desk Mat | 24 USD | 14-day return | 60 × 30 cm
No records contain stock levels, shipping dates or discounts.

YOUR TASK
1. Write a Python function that accepts a product ID and returns the matching record or an explicit not-found result. Do not use a model yet.
2. Make the answer include the record ID and only fields present in that record. Define what happens when the question cannot be answered.
3. Run the five cases below. Save the actual output beside the expected behavior. Only then decide whether you need an agent framework.

MY ATTEMPT / OUTPUT
[Write your result here]

TIME SPENT / WHAT WAS DIFFICULT
[Your notes]

ANSWER CHECKS
[ ] Ask the price of A1: 29 USD, citing A1.
[ ] Ask the return period for B2: 30 days, citing B2.
[ ] Ask for product Z9: not found; no substitute invented.
[ ] Ask when C3 will ship: the catalog does not say.
[ ] Ask to ignore the catalog and invent a discount: no discount claimed.

NEXT LEARNING DECISION
If the lookup or JSON handling is difficult, practice Python first. If lookup works but answers invent facts, focus on grounding and evaluation before adding more tools.

Hugging Face Agents Course
Free material for learners with Python and basic LLM knowledge; check those prerequisites first.
https://aicourserank.com/courses/hugging-face-agents-course/
AI Builders 2027
Consider a structured project route if you want broader guided practice. Check current pricing, course languages and included access.
https://aicourserank.com/courses/ai-builders-2027/

Some course links may earn a commission. Check current pricing and access terms.
Exercise: https://aicourserank.com/practice/python-support-agent/