Ecobolt

AI insight of the field
Analytics of fileld data
Appliance control
Threshold setting
Device detail
Email Alert
Whatsapp Alert
SMS Alert
Weather information

🌱 Inspiration

Farming today still relies heavily on manual observation and guesswork. Small-scale farmers often lack access to real-time data and intelligent tools, making it hard to respond quickly to changing environmental conditions. We wanted to bridge this digital gap by building a smart, AI-powered farming assistant that’s affordable, scalable, and farmer-friendly.

⚙️ What it does

EcoBolt is an agricultural IoT monitoring system that:

Captures real-time environmental data (temperature, humidity, light, soil conditions, nutrients)
Visualizes live and historical data through an interactive dashboard
Sends alerts via SMS/email when parameters breach critical thresholds
Offers AI-powered recommendations using IBM WatsonX
Controls appliances (pumps, lights, etc.) remotely via Bolt IoT

🛠️ How we built it

Frontend: React 18, TypeScript, Tailwind CSS, Chart.js
Backend: Supabase (PostgreSQL, Realtime, Auth), Edge Functions
Hardware: ESP32 microcontroller with soil, light, and weather sensors
AI Layer: IBM WatsonX for generating contextual farming suggestions
Notifications: Salesforce email/SMS integration with Twilio
Other APIs: OpenWeatherMap for weather, Bolt Cloud for appliance control

🚧 Challenges we ran into

Real-time synchronization between ESP32 and Supabase
Managing consistent device authentication and secure data flow
Fine-tuning WatsonX prompts to produce localized and actionable advice
Formatting email/SMS alerts reliably via Salesforce for all devices
Handling fallback when internet/cloud services were unavailable

🏆 Accomplishments that we're proud of

Fully functional production-grade dashboard with real-time updates
Successfully integrated WatsonX for smart, AI-based recommendations
Built a modular device management system with multi-device support
Enabled remote appliance control with status feedback using Bolt Cloud
Created a scalable, end-to-end system deployable on real farms

📚 What we learned

Practical use of Supabase’s real-time capabilities and Row Level Security
Efficient IoT data handling with ESP32 and multiple sensors
Best practices in AI prompt engineering for actionable results
Importance of weather-awareness and fallback systems in farming
Real-world usability factors for rural, low-connectivity areas

🚀 What's next for EcoBolt

Launching a mobile app for farmers with offline support
Training local AI models for low-connectivity, edge-based insights
Incorporating computer vision for pest/disease detection
Expanding to hydroponic and polyhouse environments
Partnering with agricultural institutions for real-world pilots

🏁 Challenge Compliance

Deploy Challenge:
EcoBolt has been successfully deployed as a full-stack application using Netlify. The project demonstrates smooth integration between frontend, backend, and real-time hardware communication — all hosted seamlessly on Netlify. The live deployment ensures reliable access and responsive performance across devices.

🔗 Live Site

Startup Challenge:
EcoBolt leverages Supabase for authentication, real-time database updates, row-level security, and serverless Edge Functions. The architecture is designed to scale with ease — from managing multiple devices per farm to handling high-frequency sensor data. This setup positions EcoBolt to scale to millions of users and IoT events with minimal overhead.
Inspirational Story:
EcoBolt was born from a personal motivation to empower farmers in rural areas with accessible technology. As someone who has seen the challenges faced in traditional agriculture, I wanted to create a tool that could bring AI and IoT directly to the fields. Built entirely using Bolt.new's prompt-based workflow, this project reflects overcoming both technical and resource limitations, proving that with the right tools and vision, innovation can be both meaningful and scalable.