Scribd
Upload
6 views1 page

Agentic AI Assignment

The assignment requires students to build an AI agent that answers customer inquiries for an online store by selecting and utilizing appropriate tools. Students must create a function to handle various questions, ensuring proper error handling and user-friendly responses. Deliverables include source code, a README file, sample inputs and outputs, and a brief document on design decisions, with optional bonus points for additional features.

Uploaded by

daddykademo
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
6 views1 page

Agentic AI Assignment

The assignment requires students to build an AI agent that answers customer inquiries for an online store by selecting and utilizing appropriate tools. Students must create a function to handle various questions, ensuring proper error handling and user-friendly responses. Deliverables include source code, a README file, sample inputs and outputs, and a brief document on design decisions, with optional bonus points for additional features.

Uploaded by

daddykademo
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Agentic AI Assignment

Submission Timeline: 5–7 Days


Level: Fresher / Final Year Students
Objective: Build a simple AI Agent that can answer customer questions for an online store by
deciding which tools to use and providing customer-friendly responses.

Problem Statement

Build an AI agent that answers customer questions about an online store. Examples: Where is
order ORD-1002? Is there a cheaper alternative to the shoes I ordered? The agent must decide
which tool(s) to use, call them in the correct order, and generate a clear answer for the customer.

Available Tools get_order(order_id) – Fetch order details. search_products(query) – Search


products using keywords. get_product(product_id) – Fetch product details.

Assignment Requirements Create a function: run_agent(question: str) -> str Identify which
tool(s) should be called based on the user's question. Chain multiple tool calls when required.
Handle invalid orders or products gracefully. Handle empty search results without guessing
information. Return customer-friendly responses instead of raw tool outputs. Do not fabricate data.

Deliverables Source Code (GitHub Repository preferred). README file explaining the approach.
Sample inputs and outputs. Short document (1–2 pages) explaining design decisions.

Bonus Points (Optional) Use any LLM provider (OpenAI, Gemini, Claude, etc.). Add logging of
tool calls. Create a simple web interface using Streamlit or Flask. Write unit tests.

Evaluation Criteria Correct tool selection – 30% Reasoning and tool chaining – 25% Error
handling – 20% Code quality and documentation – 15% User experience of responses – 10%

Submission Deadline: Submit within 5–7 days from the assignment date.
Note: You may use AI tools (ChatGPT, Gemini, Claude, Copilot, etc.) to assist in development.
However, you should clearly understand and explain your implementation during the evaluation
discussion.

You might also like