🧠 AI-Memory - Your Intelligent Digital Memory Assistant

AI-Memory is a cutting-edge web application that combines traditional content management with AI-powered querying capabilities. Save your digital content and ask intelligent questions about it using advanced vector embeddings and AI models.

🚀 Live Demo

🌐 Try AI-Memory Live

Experience the full application without any setup required! Create an account and start building your intelligent digital memory.

🌟 Key Features

🤖 AI-Powered Querying

Semantic Search: Query your saved content using natural language
Vector Embeddings: Content automatically converted to vector embeddings using Pinecone
Intelligent Responses: Powered by Google's Gemini AI for contextual answers
Source Attribution: AI responses show which of your saved items were referenced

📚 Multi-Platform Content Storage

Documents: Save text-based content and documents
YouTube Videos: Store and query video content
Social Media: Support for Twitter, LinkedIn, Instagram posts
Web Links: Save any web content with descriptions
Pinterest Pins: Visual content storage

🔐 Secure & Private

JWT Authentication: Secure user authentication system
Private Namespaces: Your content is isolated using Pinecone namespaces
Password Security: Bcrypt hashing with 12 salt rounds
CORS Protection: Configured for secure cross-origin requests

🌐 Sharing Capabilities

Public Brain Sharing: Share your memory collection with others
Secure Hash Links: Generated unique links for sharing
Read-Only Access: Shared collections are view-only for privacy

🛠️ Tech Stack

Frontend

React 18 with TypeScript
Vite for fast development and building
Tailwind CSS for responsive styling
React Router for navigation
Axios for API communication

Backend

Node.js with Express.js
TypeScript for type safety
MongoDB with Mongoose ODM
JWT for authentication
Bcrypt for password hashing

AI & Vector Database

Google Gemini AI (gemini-2.0-flash-exp model)
Pinecone vector database for embeddings
Zod for input validation

🚀 Getting Started

Prerequisites

Node.js (v18 or higher)
MongoDB database
Pinecone account and API key
Google AI Studio API key

Backend Setup

Clone the repository

git clone https://github.com/Tiger-raj/ai-memory.git
cd ai-memory/backend

Install dependencies
```
npm install
```

Environment Variables Create a .env file in the backend directory:

PORT=5000
MONGODB_URI=your_mongodb_connection_string
JWT_SECRET=your_jwt_secret_key
PINECONE_API_KEY=your_pinecone_api_key
PINECONE_INDEX_NAME=your_pinecone_index_name
PINECONE_HOST=your_pinecone_host_url
GEMINI_API_KEY=your_google_gemini_api_key
NODE_ENV=development

Start the backend server
```
npm run dev
```

Frontend Setup

Navigate to frontend directory
```
cd ../frontend
```
Install dependencies
```
npm install
```
Environment Variables Create a .env file in the frontend directory:
```
VITE_BACKEND_URL=http://localhost:5000
```
Start the development server
```
npm run dev
```
Open your browser Navigate to http://localhost:5173

📖 API Documentation

Authentication Endpoints

POST `/api/v1/signup`

Create a new user account.

{
  "username": "string (6-30 chars)",
  "password": "string (6-30 chars, uppercase, number, special char)"
}

POST `/api/v1/signin`

Sign in with existing credentials.

{
  "username": "string",
  "password": "string"
}

Content Management

POST `/api/v1/content`

Add new content to your memory.

{
  "title": "string",
  "link": "string (optional for documents)",
  "type": "youtube|twitter|pinterest|linkedin|document|link|instagram",
  "description": "string (optional, max 1000 chars)"
}

GET `/api/v1/content`

Retrieve all your saved content.

DELETE `/api/v1/content/:id`

Delete specific content item.

AI Query System

POST `/api/v1/query`

Query your content with AI.

{
  "query": "string (max 1000 chars)"
}

Response:

{
  "message": "Query processed successfully",
  "response": "AI-generated response",
  "query": "original query",
  "sourceCount": 3
}

Brain Sharing

POST `/api/v1/brain/share`

Toggle sharing of your memory collection.

{
  "share": true
}

GET `/api/v1/brain/share`

Check current sharing status.

GET `/api/v1/brain/:hash`

Access shared memory collection.

🎯 Usage Examples

Adding Content

Click "Add Content" in your dashboard
Select content type (Document, YouTube, etc.)
Enter title and optional link/description
Content is automatically processed and stored

Querying with AI

Click "Query AI" in your dashboard
Ask natural language questions about your content
Receive intelligent responses with source attribution
Example: "What are the key points from my React tutorial videos?"

Sharing Your Memory

Go to your dashboard
Toggle "Share Brain" to enable sharing
Copy the generated share link
Others can view (but not edit) your collection

🏗️ Project Structure

ai-memory/
├── frontend/
│   ├── src/
│   │   ├── components/     # Reusable UI components
│   │   ├── pages/          # Page components
│   │   ├── icons/          # Icon components
│   │   └── config.ts       # Configuration
│   ├── index.html
│   └── package.json
├── backend/
│   ├── src/
│   │   ├── controllers/    # Route handlers
│   │   ├── models/         # Database models
│   │   ├── routes/         # API routes
│   │   ├── middleware/     # Custom middleware
│   │   ├── utils/          # Utility functions
│   │   └── config/         # Database config
│   └── package.json
└── README.md

🔧 Advanced Configuration

Pinecone Vector Database

Index dimension: Configured for Gemini embeddings
Namespaces: User-specific isolation
Dynamic topK: Intelligent result count based on collection size

AI Model Configuration

Model: gemini-2.0-flash-exp
Context window: Optimized for memory querying
Response format: Conversational, 100-150 words

🌐 Deployment

Live Application

Frontend: Deployed on Vercel at https://ai-memory-five.vercel.app/
Backend: Deployed on Render with MongoDB Atlas

Backend Deployment (Render)

Set environment variables in your Render dashboard
Deploy from GitHub repository
Ensure MongoDB Atlas connection

Frontend Deployment (Vercel/Netlify)

Connect your GitHub repository
Set VITE_BACKEND_URL to your deployed backend URL
Build command: npm run build
Output directory: dist

🤝 Contributing

Fork the repository
Create a feature branch (git checkout -b feature/amazing-feature)
Commit your changes (git commit -m 'Add amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

👨‍💻 Author

Priyanshu Bajpai

GitHub: @Tiger-raj
Project: Built with passion for intelligent digital organization

🙏 Acknowledgments

Google AI for Gemini API
Pinecone for vector database services
MongoDB for database hosting
Vercel for frontend deployment
Render for backend deployment

Note: This is a demonstration project showcasing AI-powered content management. Currently, AI queries work best with document-type content. Future updates will expand AI querying to all content types.

Made with ❤️ for better digital organization

Name		Name	Last commit message	Last commit date
Latest commit History 51 Commits
backend		backend
frontend		frontend
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md

License

Tiger-raj/ai-memory

Folders and files

Latest commit

History

Repository files navigation