🧠 IntelliMeet

The Next-Generation AI-Powered Enterprise Meeting & Collaboration Platform

Features • Tech Stack • Project Structure • Getting Started

---

IntelliMeet is a secure, full-stack, enterprise-grade real-time collaboration workspace. It features advanced video conferencing via WebRTC, rich instant messaging, Kanban-based task allocation, workspace organization, push notifications, and powerful AI integrations powered by Google Gemini (providing live summaries, agenda predictions, productivity metrics, and multimodal audio transcribing).

✨ Key Features

🎥 Video & Audio Collaboration

• Real-time WebRTC: Low latency peer-to-peer and relayed video/audio routing.
• Live Transcript & Recording: Cloud-recorded meetings with live captions.
• Reactions & Screen Sharing: Express yourself and present with high frame-rate screen sharing.

🤖 AI-Powered by Google Gemini

• Auto-Summarization: Instantly generates comprehensive meeting summaries the moment a meeting ends.
• Action Item Extraction: AI automatically creates Kanban tasks from meeting transcripts.
• Productivity Analytics: Engagement metrics and meeting health scores calculated by Gemini.

📋 Team Workspace & Kanban

• Task Management: Interactive drag-and-drop Kanban boards.
• Real-time Messaging: Instant chat via Socket.io with typing indicators and online presence.
• Team Organization: Workspaces with invite links and role-based access.

🔒 Enterprise Security

• JWT & Refresh Token Rotation: Secure httpOnly cookie-based authentication.
• Data Hardening: XSS protection, NoSQL injection prevention, and strict CORS.
• Rate Limiting & Redis: Fast session handling and abuse protection.

---

🛠️ Tech Stack

Frontend (intellmeet-frontend)

• Core: React 18 (Vite), TypeScript
• Styling: Tailwind CSS, Radix UI, Lucide Icons, Framer Motion
• State & Fetching: Zustand, React Query (TanStack), Axios
• Real-time: Socket.io-client, WebRTC APIs

Backend (intellmeet-backend)

• Core: Node.js (v20+), Express.js
• Database & Cache: MongoDB Atlas (Mongoose), Redis (ioredis)
• Real-time: Socket.io v4
• AI & Cloud: Google Gemini SDK, AWS S3 (Recordings), Cloudinary (Avatars/Attachments)
• Security: bcryptjs, helmet, xss, express-mongo-sanitize

---

📁 Project Structure

This repository is structured as a monorepo containing both the frontend and backend applications as sibling directories.

IntelliMeet/
├── intellmeet-backend/       # Node.js/Express API, WebSocket server, and AI logic
│   ├── src/                  # Controllers, Models, Routes, Services, Sockets
│   ├── tests/                # Jest integration tests
│   ├── .env.example          # Backend environment variables template
│   └── package.json
│
├── intellmeet-frontend/      # React/Vite web application
│   ├── src/                  # Components, Hooks, Features, Stores, Layouts
│   ├── .env.example          # Frontend environment variables template
│   └── package.json
│
└── .gitignore                # Root gitignore

---

🚀 Getting Started

Prerequisites

• Node.js (v20 or higher)
• MongoDB instance (local or Atlas)
• Redis server (local or cloud)

1. Clone the Repository

git clone https://github.com/yasar-pathan/IntelliMeet.git
cd IntelliMeet

2. Setup the Backend

Open a new terminal window:

cd intellmeet-backend
npm install
cp .env.example .env   # Fill in your MongoDB, Redis, and Gemini API keys in .env
npm run dev            # Starts the backend server on http://localhost:5000

3. Setup the Frontend

Open a second terminal window:

cd intellmeet-frontend
npm install
cp .env.example .env   # Configure the API URL (VITE_API_URL=http://localhost:5000/api/v1)
npm run dev            # Starts the Vite dev server on http://localhost:5173

---

🏗️ Architecture Flow

graph TD
    A[Frontend React App] -->|HTTPS REST API| B(Express App Setup)
    A -->|WebSocket Connection| C(Socket.io Engine)
    
    B --> D{JWT Auth Guard}
    D -->|Valid Tokens| E[Controllers / Routing]
    
    E --> F[Services Layer]
    F -->|SDK Calls| G[Google Gemini AI]
    F -->|File Uploads| H[Cloudinary Storage]
    F -->|Recordings| I[AWS S3 Storage]
    F -->|Query / Updates| J[(MongoDB Atlas)]
    
    C -->|Handshake Auth| K[Socket Manager]
    K --> L[Presence & Chat Handler]
    K --> M[WebRTC signaling relay]
    
    L -->|Fast Cache & Append| O[(Redis Server)]
    J & O --> P[Analytics Pipeline]

Loading

---

🌐 Cloud Deployment Architecture

graph TD
    classDef client fill:#e1f5fe,stroke:#039be5,stroke-width:2px;
    classDef aws fill:#fff3e0,stroke:#ff9800,stroke-width:2px;
    classDef ec2 fill:#efebe9,stroke:#5d4037,stroke-width:2px;
    classDef external fill:#e8f5e9,stroke:#4caf50,stroke-width:2px;

    subgraph Client ["Client Layer"]
        Browser["User Browser (React App)"]:::client
    end

    subgraph AWS ["AWS Cloud Infrastructure"]
        Amplify["AWS Amplify CDN (Frontend Host)"]:::aws
        ALB["Application Load Balancer (ALB)"]:::aws
        
        subgraph EC2Instance ["EC2 Instance VM"]
            Backend["Express.js Server (Port 5000)"]:::ec2
            Redis["Local Redis (Port 6379)"]:::ec2
        end
        
        S3["Amazon S3 Bucket (Recordings)"]:::aws
    end

    subgraph External ["Managed SaaS / Database Services"]
        MongoDB["MongoDB Atlas (Persistent Data)"]:::external
        Gemini["Google Gemini AI (AI Core)"]:::external
        Cloudinary["Cloudinary CDN (Media CDN)"]:::external
    end

    %% Connections
    Browser -->|1. Fetches Static HTML/JS| Amplify
    Browser -->|2. HTTPS API / WSS WebSockets| ALB
    ALB -->|3. Routes Traffic on Port 5000| Backend
    Backend <-->|4. Local Memory Access| Redis
    Backend <-->|5. CRUD Queries| MongoDB
    Backend -->|6. AI Prompt & Transcripts| Gemini
    Backend -->|7. Image Uploads| Cloudinary
    
    %% Recording flow
    Backend -->|8. Generates Presigned URLs| S3
    Browser -->|9. Direct Upload / Download| S3

    %% Styling Link Colors
    linkStyle 0,1,2 stroke:#039be5,stroke-width:2px;
    linkStyle 3 stroke:#5d4037,stroke-width:2px;
    linkStyle 4,5,6 stroke:#4caf50,stroke-width:2px;
    linkStyle 7,8 stroke:#ff9800,stroke-width:2px;

Loading

---

Built with ❤️ for modern engineering teams.