LannzAI - Personal AI Chatbot

This is a full-stack AI chatbot application featuring a React/Vite/TypeScript frontend and a Python/Flask backend powered by a local Ollama model.

Features

• Interactive and responsive chat interface.
• Real-time AI responses generated by a local LLM (Llama 3).
• Typing indicators and smooth scrolling for a great user experience.
• AI is personalized with a custom knowledge base (my_facts.txt).

🚀 How It Works

This application operates on a full-stack architecture, allowing the frontend to communicate with a dedicated backend server that orchestrates AI interactions.

• Frontend (Client-Side):

  • Developed with React, Vite, and TypeScript for a blazing-fast development experience and robust code.
  • Styled using Tailwind CSS for a utility-first approach to design.
  • Manages the chat history, user input, and displays messages with real-time UI updates
  • Communicates with the backend API to send user messages and receive AI responses.

• Backend (Server-Side):

  • Built with Python and Flask, providing a lightweight and efficient API endpoint.
  • Acts as an intermediary, receiving user queries from the frontend.
  • Forwards these queries to the local AI model (Ollama) along with a custom "system prompt" and your personal facts.
  • Processes the AI's response and sends it back to the frontend.

• AI Core (Local LLM):

  • Powered by Ollama, which enables running large language models like Llama 3 directly on your computer.
  • This ensures privacy, speed, and completely free AI inference after the initial model download.
  • The AI model uses the provided my_facts.txt as its context to generate relevant and personalized answers.

---

🛠️ Technologies Used

Frontend

• React: A JavaScript library for building user interfaces.
• Vite: A fast build tool that significantly improves frontend development experience.
• TypeScript: A typed superset of JavaScript that compiles to plain JavaScript.
• Tailwind CSS: A utility-first CSS framework for rapidly building custom designs.
• Lucide React: Beautiful, open-source icons for React.

Backend

• Python: The versatile programming language for server logic.
• Flask: A micro web framework for Python, used for creating the API.
• Flask-CORS: Enables Cross-Origin Resource Sharing for seamless frontend-backend communication.
• OpenAI Python Client: Used to interact with the local Ollama API (which mimics OpenAI's API structure).

AI / LLM

• Ollama: A powerful tool for running large language models locally.
• Llama 3: A state-of-the-art open-source LLM from Meta, serving as the chatbot's brain.