Skip to content

Latest commit

 

History

History
141 lines (106 loc) · 5.88 KB

README.md

File metadata and controls

141 lines (106 loc) · 5.88 KB

🌐🤖 Mult-AI: Your All-in-One AI Query Solution

Mult-AI is an innovative platform that combines powerful AI capabilities to make querying SQL databases and PDF documents effortless. With Mult-AI, interact seamlessly with databases through natural language and get answers based on PDF content—all within a single, easy-to-use interface!

To know more about the app's functionality head over to this README.md file

Table of Contents

Technologies Used

Frontend

  • React: A JavaScript library for building user interfaces.
  • Tailwind CSS: A utility-first CSS framework for rapid UI development.
  • Marked: A markdown parser and compiler for converting markdown text to HTML.
  • Axios: For making API requests from the frontend to the backend.

Backend

  • FastAPI: A modern, fast web framework for building APIs with Python 3.7+ based on standard Python type hints.
  • Markdown2: A Python library to convert markdown text to HTML on the backend.
  • Uvicorn: An ASGI server for serving FastAPI applications in production.
  • AWS RDS,EC2: Cloud based data storage for getting public endpoint.

API

  • Google Gemini API: Used for extracting texts from pdf and generating natural language response.
  • Groq API: Used to create and process sql chain and generate response in natural language with respect to the user's question based on database.

Authentication: Clerk

Deployment: Vercel (Frontend), Render (Backend)

Features

  • \🗃️ Multi SQL Query: Connect to SQL databases, input natural language queries, and receive AI-generated SQL responses instantly.
  • 📄 Multi PDF Chat: Upload PDFs, ask questions, and get AI-generated answers based on the content of your documents.
  • 🌐🇮🇳 Hinglish Support: Communicate with the AI in Hinglish (a blend of Hindi and English), creating a more natural experience for multilingual users. Powered by the Deep Translator library, Mult-AI bridges language gaps effortlessly!
  • 🖋️ HTML Formatted Output: Enjoy responses with proper HTML formatting, ensuring clean and readable output without the hassle of markdown symbols.
  • 📱 Responsive Design: Built with Tailwind CSS, the UI is optimized for a smooth experience on both desktop and mobile devices.

Unlock a new way to query and interact with your data, all in one intuitive platform! 🚀

Project Workflow

  1. Frontend Development:

    • The frontend is built with React. User queries are sent from the frontend to the FastAPI backend using Axios.
    • Responses from the backend are rendered using dangerouslySetInnerHTML in React, which allows HTML-formatted responses.

  2. Backend Development:

    • The backend is built with FastAPI. Upon receiving a user query, the backend interacts with the AI model to generate a response.
    • The response from the AI model is processed with markdown2 to convert markdown syntax to HTML for proper formatting.

  3. Markdown Conversion:

    • If the AI response contains markdown, it is converted to HTML using marked in the frontend.
    • The final response with HTML formatting is then sent to the frontend.

  4. Deployment:

    • The backend is served using Uvicorn and can be deployed on platforms like Render or AWS.
    • The frontend can be deployed on Vercel or Netlify for easy hosting and scalability.

Installation

Prerequisites

  • Node.js and npm
  • Python 3.7+
  • Virtual environment (optional but recommended)

Backend Setup

  1. Clone the repository and navigate to the backend directory: 
    git clone https://github.com/SahiLmb/AI-Sql-query-gen.git
cd AskQL/backend

Install dependencies and activate the virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows, use `venv\Scripts\activate`
pip install -r requirements.txt

Run the backend server:

uvicorn app:app --reload

Frontend Setup

Navigate to the frontend directory and install dependencies:

cd AskQL/frontend/sql-gen-ai
npm install

Run the frontend server:

npm run dev

Setting up Environment Variables

Create a .env.local file in the root directory in frontend and add the following environment variables:

NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY=your_publishable_key
CLERK_SECRET_KEY=your_secret_key

NEXT_PUBLIC_CLERK_SIGN_IN_URL=/sign-in
NEXT_PUBLIC_CLERK_SIGN_UP_URL=/sign-up

NEXT_PUBLIC_CLERK_AFTER_SIGN_IN_URL=/dashboard
NEXT_PUBLIC_CLERK_AFTER_SIGN_UP_URL=/dashboard

NEXT_PUBLIC_API_URL=your_api_url

Usage

Accessing Multi SQL Query

  • Navigate to the Multi SQL Query Feature: Use the "Multi SQL Query" link in the header to go to the SQL querying section.
  • Upload SQL Database: Upload your SQL database file to start querying.
  • Ask Questions: Enter questions in natural language. The AI model will convert these into SQL queries and fetch answers directly from your database.

Accessing Multi PDF Chat

  • Navigate to the Multi PDF Chat Feature: Use the "Multi PDF Chat" link in the header.
  • Upload PDFs: Upload one or more PDF files.
  • Ask Questions: Ask questions about the content in the PDFs. The AI will analyze the content and provide answers based on the information within the uploaded documents.

Once both servers are running, navigate to http://localhost:3000 in your browser to access the application. Type a query into the chat input box, and receive formatted responses from the AI.

Contributing

  • Fork the repository.
  • Create a feature branch (git checkout -b feature/AmazingFeature).
  • Commit your changes (git commit -m 'Add some AmazingFeature').
  • Push to the branch (git push origin feature/AmazingFeature).
  • Open a pull request.