BeejBandhu

Inspiration

BeejBandhu was inspired by the everyday struggles of small and marginal farmers who depend on uncertain weather, local advice, and guesswork for critical farming decisions. Our team realized that while farmers have access to smartphones and government schemes, they often lack a unified, reliable advisory system that connects them with accurate information on seeds, soil, weather, and markets. This motivated us to build BeejBandhu — a smart, AI-driven crop advisory companion designed to empower farmers with data-backed guidance while keeping it simple and accessible.

What We Learned

During this project, we learned how deeply agriculture depends on data — from soil health and rainfall patterns to pest outbreaks and pricing trends. We explored how AI, IoT, and satellite data can work together to provide timely insights to farmers in rural India. We also learned the importance of designing for low internet connectivity and local languages, ensuring inclusivity for farmers with limited digital literacy. Beyond technology, we understood the value of empathy — that innovation in agri-tech must start from the ground realities of the farming community.

How We Built It

We approached BeejBandhu as a modular, scalable system combining data analytics, AI, and accessible interfaces:

1. Data Integration: Aggregated soil, weather, and crop data from open APIs (IMD, ISRO, AgriStack).
2. AI Crop Advisory: Used predictive models to recommend suitable crops, fertilizers, and pest control measures based on soil health, weather forecasts, and historical yield data.
3. Pest & Disease Detection: Integrated a photo-based detection model (CNN) allowing farmers to upload crop images and get instant diagnosis and remedy suggestions.
4. Weather & Market Alerts: Implemented real-time alerts for rainfall, temperature changes, and market prices to help farmers make timely decisions.
5. Multilingual Chatbot Interface: Built an intuitive chatbot accessible via web or mobile that provides personalized, easy-to-understand advice.
6. Scheme Integration: Added a rule-based DSS module linking eligible government schemes (PM-KISAN, DAJGUA, Jal Jeevan Mission) to farmers based on their land and crop profile.

The system is hosted on the MERN stack, with a responsive UI built using React and Tailwind CSS, a Node.js backend, MongoDB for storage, and AI models hosted via Python APIs.

Challenges We Faced

• Data inconsistency: Agricultural datasets vary in format and accuracy, requiring careful cleaning and validation.
• Rural connectivity: Ensuring fast load times and offline features for low-network areas was a major challenge.
• Model reliability: Calibrating AI models for diverse soil types and local climates required iterative fine-tuning.
• User experience: Simplifying tech-heavy insights into visuals and short, regional-language

Built With

• analytics
• and
• and-advisory-records.-ai/ml-components:-python-for-ai-and-data-analysis
• and-javascript-(es6)-for-interactivity-and-logic-handling.-backend:-node.js-for-scalable-server-side-operations-and-express.js-for-creating-secure-and-efficient-rest-apis.-database:-mongodb-as-a-nosql-database-for-storing-farmer-data
• and-market-price-apis-for-dynamic-crop-pricing-information.-visualization-&-tools:-mapbox/leaflet.js-for-interactive-maps-and-resource-visualization
• and-opencv-for-image-based-disease-detection.-apis-&-integrations:-weather-apis-(imd/openweather)-for-real-time-weather-data
• chart.js/d3.js
• crop-insights
• dashboards
• for
• frontend:-reactjs-for-building-an-interactive-and-responsive-user-interface
• graphical
• isro/bhuvan-apis-for-soil-and-satellite-analysis
• tailwind-css-for-clean-and-mobile-friendly-design
• tensorflow/pytorch-for-crop-and-pest-prediction-models