Scribd
Upload
32 views2 pages

AI Intern Assignment

The assignment requires the creation of a Retrieval-Augmented Generation (RAG) Chatbot using Python and LangChain, with three main tasks: loading a dataset, setting up the RAG pipeline, and building the chatbot. Students must document their code, save sample questions and responses, and submit all work via a GitHub repository link shared through email. Bonus points are available for deploying the chatbot on Streamlit.

Uploaded by

Misa Amane
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
32 views2 pages

AI Intern Assignment

The assignment requires the creation of a Retrieval-Augmented Generation (RAG) Chatbot using Python and LangChain, with three main tasks: loading a dataset, setting up the RAG pipeline, and building the chatbot. Students must document their code, save sample questions and responses, and submit all work via a GitHub repository link shared through email. Bonus points are available for deploying the chatbot on Streamlit.

Uploaded by

Misa Amane
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
You are on page 1/ 2

Assignment

Instructions:
This assignment contains a single question with three key tasks. Please ensure you follow all
guidelines and adhere to coding standards.

● Deadline: To be announced via email.


● Submission: Upload all relevant files to GitHub and share the repository link via email
before the deadline.
● Comments: Include comments in your code wherever necessary.
● Citation: Properly cite any external sources used.

Objective:

This assignment will assess your skills in Retrieval-Augmented Generation (RAG), Python, and
LangChain for building a Chat application. The task is to create a RAG-based Chatbot that uses
external data sources to answer questions accurately and contextually.

Assignment Overview:

Task 1: Data Loading

Load any dataset of your choice. Ensure it is formatted in a way that can be easily queried (e.g.,
CSV, JSON, or any document format). This data will serve as the knowledge base for your RAG
Chatbot.

Task 2: Set Up RAG with LangChain

Use LangChain’s tools to set up a Retrieval-Augmented Generation pipeline. Implement


LLMChain or similar components in LangChain for retrieval and question answering. You can
refer to the LangChain documentation to guide you through setting up the RAG pipeline.

Task 3: Build the Chatbot

Using the RAG pipeline, build a Chatbot that can answer questions from users based on the
data loaded in Task 1. Ensure the responses are accurate and utilize retrieval from the
knowledge base.
Requirements:

● Code Documentation: Ensure all classes and functions are well-documented and
adhere to Python coding standards.
● Response File: Save a sample set of questions and chatbot responses in a .txt,
.pdf, or .xlsx file.
● GitHub Repository: Push all code (including notebooks if using Google Colab) and
relevant files to a GitHub repository, then share the link.

Bonus Points:

Earn additional points if you deploy your Chatbot on Streamlit and provide the application link in
the submission email.

Deliverables:

1. Python Code
○ Submit either .ipynb (Google Colab) or .py files with proper comments and
documentation.
2. Sample Questions and Chatbot Responses
○ Save these in a .txt, .pdf, or .xlsx file.
3. GitHub Repository
○ Push all code, notebooks, and relevant files to the repository.
4. Optional Streamlit Link (Bonus)
○ If hosted, include the Streamlit application link in your submission email.

Submission:
● You will have to email the Github repo link and other relevant links/files to
[email protected]
● You may use resources such as GitHub Copilot, ChatGPT, or other tools for assistance.
Ensure all resources are correctly cited.

Note: For any questions or guidance, please contact us at [email protected] OR


[email protected] .

You might also like