KhetSaathiAI is a state-of-the-art agricultural diagnostic and intelligence platform designed to empower farmers with AI-driven insights. It provides plant disease detection, weather forecasts, real-time mandi prices, and personalized crop recommendations.
|
|
|
|
The project is organized as a monorepo:
/frontend: React + Vite application. It includes a built-in Express proxy server (server.ts) for handling SMTP (emails) and secure API communications./backend: Python FastAPI service for AI model orchestration, agricultural data processing, and predictive analytics.
- Node.js (v18+)
- Python (3.9+)
- Firebase Account (for Authentication and Database)
- API Keys: Groq (Llama 3), Gemini (Vision), and OpenWeatherMap (optional/integrated).
- Copy the root
.env.exampleto a new file named.env.cp .env.example .env
- Open
.envand fill in your API keys and credentials. - The system is designed to read the
.envfrom the root or sub-directories depending on your environment setup.
cd frontend
npm install
npm run devThe frontend will be available at http://localhost:5173.
cd backend
pip install -r requirements.txt
python app.pyThe backend API will run on http://localhost:8000.
- 🌿 AI Agricultural Assistant: A specialized LLM (Llama 3 via Groq) tailored for farming queries.
- 🔍 Disease Diagnosis: Multi-modal AI (Gemini Vision) to detect crop diseases from images.
- 📊 Real-time Mandi Prices: Integration with market data for crop price forecasting.
- 🌦️ Smart Weather Updates: Localized weather alerts specifically for agricultural needs.
- 🗣️ Voice Support: Integrated Speech-to-Text and Text-to-Speech for farmer accessibility.
- 🌍 Multi-lingual: Seamless support for English and Hindi (हिन्दी).
- Frontend: React, Vite, Tailwind CSS, Motion (Framer Motion), Lucide Icons.
- Backend: Python, FastAPI, Groq SDK, Google Generative AI, Pandas, Scikit-learn.
- Database/Auth: Firebase.
- Proxy/Email: Node.js, Express, Nodemailer.
Unnecessary development scripts and temporary folders have been removed to ensure a clean repository structure for production and GitHub sharing.
This project is licensed under the MIT License - see the LICENSE file for details.