Bangla EduAgent 🎓

🔥 ⚠️ IMPORTANT: Requires Paid Google Gemini Pro API Key ⚠️ Bangla EduAgent uses advanced Gemini 3 Pro / 2.5 Flash models with Google Search grounding.

Get your paid key: Google AI Studio

AI-Powered Personalized Learning Path Generator in Bengali

Bangla EduAgent is a sophisticated educational web application designed to bridge the language gap in technical and general education. By leveraging Google's Gemini 3 Pro and Gemini 2.5 Flash models, it dynamically generates personalized courses, quizzes, and learning materials in the Bengali language based on user-defined topics.

🚀 Project Purpose

The primary goal of this project is to democratize access to high-quality education for Bengali speakers. Traditional MOOCs (Massive Open Online Courses) are predominantly in English, creating a barrier for millions of learners. Bangla EduAgent acts as an intelligent tutor that:

1. Translates & Adapts: Instantly creates content on complex topics (e.g., "Quantum Physics", "React JS") in Bengali.
2. Personalizes: Generates a syllabus tailored to the topic.
3. Validates: Uses Google Search Grounding to ensure facts are up-to-date and provides citations.

🛑 The Challenge

While information is abundant, accessibility is not distributed equally:

• Language Barrier: Tech documentation and advanced tutorials are almost exclusively in English.
• Static Content: Traditional courses become outdated quickly.
• Lack of Context: Direct translation tools (like Google Translate) often fail to capture the educational context or technical nuance required for learning.

Bangla EduAgent solves this by generating fresh, context-aware, grounded educational content on demand.

🛠️ Prerequisites & Required Tools

Before running the project, ensure you have the following installed:

• Node.js (v18.0.0 or higher)
• npm (Node Package Manager) or yarn
• Google AI Studio API Key (Required for Gemini models)

⚙️ Setup Instructions

Follow these steps to run the project locally:

1. Clone the Repository

   git clone https://github.com/your-username/bangla-eduagent.git
   cd bangla-eduagent

2. Install Dependencies

   npm install

3. Configure Environment Variables Create a file named .env.local in the root directory.

   touch .env.local

4. Add API Key Open .env.local and add your Google Gemini API key:

   API_KEY=your_actual_api_key_here

   Note: You must use a paid tier project or a valid AI Studio key to access Gemini 3 Pro and Search Grounding.

5. Start the Application

   npm start

   The app should now be running at http://localhost:3000 (or the port specified by your bundler).

💡 Solution Overview

The application follows a client-side AI architecture. It communicates directly with Google's GenAI API from the browser to minimize latency for this prototype phase.

• Syllabus Agent (Gemini 2.5 Flash): Responsible for high-speed planning. It breaks down a topic into a structured JSON syllabus.
• Content Agent (Gemini 3 Pro): Responsible for deep reasoning and content generation. It writes the markdown content for each module and generates relevant quizzes.
• State Management: Uses React Hooks (useCourseManager) and LocalStorage to persist user progress and courses offline.

✨ Key Features

• Generative Course Creation: Type any topic, and get a full course structure in seconds.
• Voice Input: Integrated Web Speech API allows users to search topics via voice commands.
• Search Grounding: Content is backed by real-time Google Search results with source links.
• Interactive Quizzes: AI-generated multiple-choice questions with detailed explanations.
• Smart Preloading: Background fetching of the "next module" ensures zero wait time between lessons.
• Progress Tracking: Dashboard with visual charts and "Certified" badges upon completion.
• Bengali Typography: Optimized fonts (Noto Sans Bengali) for readability.

🏗️ Architecture Diagrams

1. Current Architecture (Client-Side MVP)

This version is optimized for hackathons and quick deployment, leveraging React and direct Gemini API calls.

graph TD
    User((User))
    
    subgraph ClientSide ["Client Side (React)"]
        UI[User Interface]
        Voice[Voice Input]
        Store[LocalStorage]
        
        subgraph LogicLayer ["Logic Layer"]
            Mgr[Course Manager Hook]
            Preloader[Background Preloader]
        end
    end
    
    subgraph GoogleCloud ["Google Cloud / Gemini API"]
        Router{Model Router}
        Flash["Gemini 2.5 Flash<br>(Syllabus & Speed)"]
        Pro["Gemini 3 Pro<br>(Content & Reasoning)"]
        Search["Google Search<br>Grounding"]
    end

    User --> UI
    Voice --> UI
    UI --> Mgr
    Mgr <--> Store
    Mgr -->|"Generate Syllabus"| Flash
    Mgr -->|"Generate Content"| Pro
    Preloader -->|"Fetch Next Module"| Pro
    
    Flash <--> Search
    Pro <--> Search
    
    Flash -->|"JSON Structure"| Mgr
    Pro -->|"Markdown + Quiz"| Mgr

Loading

2. Future Architecture (Scalable Full-Stack)

This is the target architecture designed to support 10,000+ concurrent users, authentication, and heavy processing.

graph TD
    subgraph "Frontend (React)"
        Client[Web Client]
    end

    subgraph "Backend (Nest.js)"
        Auth[Auth Guard]
        API[API Gateway]
        Queue[Job Queue]
        GeminiSvc[Gemini Service]
    end

    subgraph "Data Layer"
        Mongo[(MongoDB)]
        Redis[(Redis Cache)]
    end

    subgraph "External AI"
        GoogleAI[Google Gemini API]
    end

    Client -- "JWT Auth" --> Auth
    Auth --> API
    API --> Mongo
    API -- "Heavy Tasks" --> Queue
    Queue --> GeminiSvc
    GeminiSvc --> GoogleAI
    GeminiSvc -- "Cache Result" --> Redis

Loading

🌍 Project Impact

• Accessibility: Removes language barriers for technical skill acquisition.
• Engagement: Gamified elements (quizzes, confetti, progress bars) keep learners motivated.
• Scalability: The AI-driven model means the platform can teach any subject without needing human authors to write the content manually.

🚀 Future Plans & Roadmap

We are transitioning from a client-side prototype to a robust full-stack application.

1. Backend Migration: Move logic to Nest.js.
2. Database: Implement MongoDB for user accounts and cloud persistence.
3. Authentication: Add Login/Signup functionality.
4. Admin Portal: Tools for monitoring usage and managing users.
5. Social Sharing: Share course certificates on social media.