Local RAG Chatbot for Resume Q&A

A simple project to run a local Large Language Model (LLM) and ask questions about your resume. This project is designed to showcase how you can use Retrieval-Augmented Generation (RAG) techniques with Ollama to privately query your own documents.

Features

• 100% local: No data leaves your machine
• Uses Ollama's Llama3 model for Q&A
• Embeds and retrieves information from your PDF resume
• Interactive command-line interface
• Easy setup with a single script

How It Works

1. Loads your resume from a PDF file
2. Splits the document into manageable text chunks
3. Generates vector embeddings for each chunk
4. Stores embeddings in a FAISS vector database
5. Accepts your questions, retrieves relevant info, and generates answers using Llama3

Setup

1. Clone this repository

   git clone https://github.com/lucasbonanni/local-resume-chatbot
   cd local-resume-chatbot

2. Add your resume
   • Place your resume.pdf in the project directory.
3. Run the initialization script

   chmod +x init.sh
   ./init.sh

4. Start the chatbot

   source venv/bin/activate
   python main.py

Example

Hello! I am your local document assistant with Ollama. Let's get started.
Loading the necessary libraries...
Loaded 5 documents from resume.pdf.
The system is ready for Q&A. You can start asking questions.

Please enter your question (or type 'exit' to quit): What programming languages do I know?

Response: According to your resume, you are proficient in Python, JavaScript, and SQL. You also have experience with TypeScript and Java.

Requirements

• Python 3.8+
• Ollama installed locally
• The following Python packages (installed by init.sh):
  • langchain
  • faiss-cpu
  • pypdf
  • ollama

---

This project demonstrates how modern AI can run privately and efficiently on your own computer. Perfect for demos, learning, or personal use!