Company Specialized Chatbot

---

Overview

This is a sophisticated, personalized chatbot application powered by a FastAPI backend and a modern, responsive frontend. It features Retrieval-Augmented Generation (RAG) for knowledge-based responses and ultra-low latency inference via Groq LPU and Llama 3.

---

Features

• Modern UI: Clean, mobile-friendly, responsive chat interface built using Vanilla HTML/CSS.
• RAG-based Intelligence: Combines local document retrieval with LLM response generation.
• FastAPI Backend: Async Python backend for low-latency interactions.
• High Performance: Powered by Groq's LPU and Meta's Llama 3.
• Modular Design: Easy to extend and maintain with separate layers for parsing, prompting, and vector DB.

---

Tech Stack

Layer	Technologies
Frontend	HTML, CSS, JavaScript
Backend	Python, FastAPI, Uvicorn, Pydantic, Groq (Llama 3), PyMuPDF, ChromaDB
Tools	uv (Python package management), make, streamlit

---

Project Structure

personalized-chatbot/
├── backend/
│   ├── src/                # FastAPI app, routing and orchestration logic
│   ├── database/           # Vector store (ChromaDB) integration
│   ├── parser/             # Document parsing logic
│   ├── prompts/            # Prompt templates
│   ├── data/               # ChromaDB persistent storage
│   ├── .env.example        # Environment config sample
│   ├── load_documents.py   # Load files to vector store
│   ├── pyproject.toml      # Backend deps (uv format)
│   ├── uv.lock             # Lockfile
│   └── requirements.txt    # Backend deps (pip format)
├── frontend/              
│   ├── index.html          # Main UI
│   ├── chatbot.js          # Core chat logic
│   ├── script.js           # Additional behavior
│   ├── chatbot.css         # Chat styles
│   └── styles.css          # General styles
├── .gitignore             
├── .python-version        
├── Makefile               
├── .env                   # API keys and environment values
└── README.md              

---

Getting Started

Prerequisites

• uv (Astral) - Fast Python package manager
• make utility (available by default on Linux/macOS)

To install uv:

Linux/macOS

curl -Ls https://astral.sh/uv/install.sh | sh

Windows

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

---

Setup Instructions

Step	With make	Without make
Clone Repository	git clone https://github.com/devrahulbanjara/personalized-chatbot	Same as left
	cd personalized-chatbot
Install Dependencies	make install	cd backend && uv sync
Load Documents	make run-loaddoc	cd backend && source .venv/bin/activate && streamlit run load_documents.py
Run Backend Server	make run-backend	cd backend && source .venv/bin/activate && python -m src.main
Open Frontend	Open frontend/index.html in browser	Same as left

Note: Loading documents to the knowledge base is a mandatory step before using the chatbot.

---

Sample API Query with CURL

Once the backend is running:

curl -X 'POST' \
  'http://localhost:8000/api/chat' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "query": "Where does KP Oli live ?",
  "session_id": "some_random_string"
}'

---

Makefile Commands

Command	Description
make install	Install all backend dependencies using uv
make run-backend	Start FastAPI backend
make run-loaddoc	Load documents (starts backend and streamlit loader)
make clean	Remove virtual environment and reset state
make help	Show all make commands

---

Contributing

1. Fork the repository
2. Create a feature branch: git checkout -b feature/my-feature
3. Commit your changes: git commit -m "Add my feature"
4. Push to the branch: git push origin feature/my-feature
5. Open a pull request

---

License

This project is licensed under the GNU GENERAL PUBLIC LICENSE.