🚀 MSME Mitr - AI-First MSME Advisory System

🌟 Overview

MSME Mitr is an AI-powered multilingual advisory system designed to help Micro, Small, and Medium Enterprises (MSMEs) in India discover and apply for government schemes. Built with a mobile-first approach, it provides personalized recommendations through an intelligent chatbot interface supporting 12 Indian languages.

🎯 Key Features

• AI-First Chatbot - Intelligent conversational interface powered by OpenRouter (Claude, GPT-4, Llama)
• Voice Input - Deepgram-powered speech-to-text in multiple Indian languages
• Offline Mode - Works without internet using cached scheme data
• Multi-Language Support - 12 Indian languages including Hindi, Tamil, Telugu, Bengali
• Real-time Streaming - Smooth streaming responses for better UX
• Smart Recommendations - Personalized scheme suggestions based on business profile
• Mobile-First Design - Optimized for rural users with poor connectivity

🚀 Quick Start

Prerequisites

• Node.js 18+ and npm
• API Keys (optional but recommended):
  • OpenRouter API Key - For AI responses
  • Deepgram API Key - For voice transcription

Installation

1. Clone the repository

git clone https://github.com/yourusername/msme-mitr.git
cd msme-mitr

2. Install dependencies

npm install

3. Set up environment variables

cp .env.local.example .env.local

Edit .env.local and add your API keys:

# OpenRouter API Configuration
OPENROUTER_API_KEY=your_openrouter_api_key_here

# Deepgram API Configuration (for voice recording)
DEEPGRAM_API_KEY=your_deepgram_api_key_here

# Model Selection (choose one)
OPENROUTER_MODEL=anthropic/claude-3-haiku  # Fast & affordable
# OPENROUTER_MODEL=anthropic/claude-3-sonnet
# OPENROUTER_MODEL=openai/gpt-3.5-turbo
# OPENROUTER_MODEL=meta-llama/llama-3-70b-instruct

# Rate Limiting
MAX_REQUESTS_PER_MINUTE=20
MAX_REQUESTS_PER_DAY=500

4. Run the development server

npm run dev

5. Open in browser

http://localhost:3000

🏗️ Architecture

Tech Stack

• Frontend: Next.js 14 (App Router), TypeScript, Tailwind CSS
• UI Components: shadcn/ui, Radix UI
• AI Integration: OpenRouter API (multi-model support)
• Voice: Deepgram STT
• State Management: React Hooks, Context API
• Streaming: Vercel AI SDK

Project Structure

msme-mitr/
├── app/
│   ├── api/
│   │   ├── chat/          # Chat endpoints
│   │   ├── voice/         # Voice transcription
│   │   └── health/        # Health check
│   ├── page.tsx           # Home page
│   └── layout.tsx         # Root layout
├── components/
│   ├── mobile/            # Mobile-optimized components
│   │   └── ChatInterfaceStream.tsx
│   ├── ui/                # Reusable UI components
│   └── layouts/           # Layout components
├── services/
│   ├── ai/                # AI services
│   │   ├── openRouterService.ts
│   │   └── contextBuilder.ts
│   ├── chat/              # Chat management
│   │   └── conversationManager.ts
│   ├── schemes/           # Scheme data service
│   │   └── schemeDataService.ts
│   ├── voice/             # Voice services
│   │   └── deepgramService.ts
│   ├── language/          # Multi-language support
│   │   └── languageService.ts
│   └── offline/           # Offline mode
│       └── offlineService.ts
├── hooks/                 # Custom React hooks
├── types/                 # TypeScript definitions
├── data/                  # Static data
│   └── schemes.json       # Government schemes database
└── public/                # Static assets

🎨 Features in Detail

1. AI-Powered Chat Interface

The chat interface uses streaming responses for a smooth experience:

• Multiple AI Models: Choose between Claude, GPT-4, Llama, and more
• Context-Aware: Maintains conversation history and user profile
• Token Optimization: Intelligent context management for cost efficiency
• Fallback Mode: Works without API keys using cached responses

2. Voice Input with Deepgram

Advanced voice recording features:

• Multi-Language STT: Supports 12 Indian languages
• Auto-Detection: Automatically detects spoken language
• Noise Cancellation: Built-in echo cancellation and noise suppression
• Visual Feedback: Real-time recording indicator with duration

3. Offline Mode

Comprehensive offline functionality:

• Automatic Detection: Detects network status changes
• Cached Responses: Pre-generated answers for common questions
• LocalStorage: Persists scheme data for offline access
• Sync on Reconnect: Auto-syncs when back online

4. Multi-Language Support

Complete localization for Indian users:

Supported Languages:
- English (en)
- हिन्दी (hi)
- தமிழ் (ta)
- తెలుగు (te)
- বাংলা (bn)
- मराठी (mr)
- ગુજરાતી (gu)
- ಕನ್ನಡ (kn)
- മലയാളം (ml)
- ਪੰਜਾਬੀ (pa)
- ଓଡ଼ିଆ (or)
- অসমীয়া (as)

5. Scheme Management

Intelligent scheme recommendation system:

• 11 Government Schemes: Curated database of MSME schemes
• Smart Filtering: Based on business type, location, and eligibility
• Token Budget: Optimizes context for API limits
• Personalization: Learns from user profile and conversation

📱 Mobile Optimization

Designed for rural Indian users:

• Large Touch Targets: 48px minimum for easy interaction
• Safe Area Padding: Accounts for device notches and home indicators
• Responsive Design: Adapts to all screen sizes
• Low Bandwidth: Optimized for 2G/3G networks
• Progressive Enhancement: Core features work without JavaScript

🔧 Configuration

Environment Variables

Variable	Description	Required	Default
OPENROUTER_API_KEY	OpenRouter API key for AI responses	No	-
DEEPGRAM_API_KEY	Deepgram API key for voice transcription	No	-
OPENROUTER_MODEL	AI model to use	No	anthropic/claude-3-haiku
MAX_REQUESTS_PER_MINUTE	Rate limit per minute	No	20
MAX_REQUESTS_PER_DAY	Rate limit per day	No	500
ENABLE_DEBUG_LOGS	Enable debug logging	No	false

Model Selection

Available models through OpenRouter:

Model	Speed	Cost	Best For
anthropic/claude-3-haiku	Fast	Low	General queries
anthropic/claude-3-sonnet	Medium	Medium	Complex tasks
anthropic/claude-3-opus	Slow	High	Advanced reasoning
openai/gpt-3.5-turbo	Fast	Low	Quick responses
openai/gpt-4-turbo	Medium	High	Complex analysis
meta-llama/llama-3-70b	Medium	Low	Open-source option

🧪 Testing

# Run development server
npm run dev

# Build for production
npm run build

# Start production server
npm start

# Run linting
npm run lint

# Type checking
npm run type-check

📊 Scheme Database

The system includes 11 government schemes:

1. PMEGP - Prime Minister's Employment Generation Programme
2. CGTMSE - Credit Guarantee Fund Trust for MSEs
3. MUDRA - Micro Units Development & Refinance Agency
4. Stand Up India - Loans for SC/ST and Women entrepreneurs
5. 59 Minute Loan - Quick MSME loans
6. Udyam Registration - Official MSME registration
7. GeM - Government e-Marketplace
8. MSME Samadhaan - Delayed payment resolution
9. Digital MSME - Digital adoption scheme
10. SFURTI - Cluster development scheme
11. ZED - Zero Defect Zero Effect certification

🤝 Contributing

We welcome contributions! Please follow these steps:

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

Development Guidelines

• Follow TypeScript best practices
• Use meaningful variable and function names
• Add comments for complex logic
• Ensure mobile responsiveness
• Test offline functionality
• Optimize for low bandwidth

🔒 Security

• API keys are never exposed to the client
• Rate limiting prevents abuse
• Input sanitization for all user inputs
• Secure headers with Next.js
• CORS properly configured

📈 Performance

Optimizations implemented:

• Edge Runtime: For streaming endpoints
• Token Optimization: Smart context management
• Lazy Loading: Components loaded as needed
• Image Optimization: Next.js Image component
• Code Splitting: Automatic with Next.js
• Caching: LocalStorage for offline mode
• CDN Ready: Static assets optimized

🎯 Roadmap

• WhatsApp Integration
• SMS-based queries
• Document upload and OCR
• Application tracking
• Video tutorials
• Regional language voice output
• Analytics dashboard
• Admin panel
• Scheme updates via CMS
• Push notifications

📜 License

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

🙏 Acknowledgments

• OpenRouter for multi-model AI access
• Deepgram for speech-to-text
• Vercel for hosting and AI SDK
• shadcn/ui for beautiful components
• Government of India for MSME scheme information

💬 Support

For support, please:

• Open an issue on GitHub
• Contact the development team
• Check the documentation

🌐 Links

• Live Demo
• Documentation
• API Reference

---

Made with ❤️ for Indian MSMEs

Empowering 6.3 Crore MSMEs across India