Inspiration
AI In Farms was inspired by the agricultural challenges faced by farmers across Africa — especially the difficulty of detecting crop diseases early, monitoring large farms efficiently, and accessing intelligent farming support without expensive infrastructure.
We wanted to build an AI-powered agricultural intelligence platform that combines futuristic operational AI systems with farmer-friendly accessibility through WhatsApp and mobile-first experiences. The goal was to create something that feels like a real agricultural operating system rather than just a simple demo.
What it does
AI In Farms is an AI-powered agricultural intelligence platform that helps farmers:
- Detect crop diseases from uploaded crop images
- Assess agricultural risks and severity levels
- Generate AI-powered treatment recommendations
- Simulate operational AI workflows visually
- Deliver recommendations through a WhatsApp-style interface
- Monitor AI workflows and analytics through operational dashboards
The platform also introduces workflow orchestration visualization, real-time operational intelligence simulations, and scalable infrastructure concepts designed for future drone-assisted crop monitoring.
How we built it
We built AI In Farms using a frontend-first architecture for the MeDo challenge, while structuring the backend architecture separately for future scalability.
Frontend
The frontend was built to deliver:
- Cinematic UI/UX
- Glassmorphism operational dashboards
- AI workflow visualizations
- Animated intelligence pipelines
- WhatsApp interaction simulation
- Real-time analytics simulations
The frontend stack includes:
- React
- Advanced animations
- Modern responsive UI systems
- Operational AI-inspired design patterns
Backend Architecture
Although the MeDo submission focuses primarily on the frontend experience, the backend system architecture was carefully designed around:
- Event-driven orchestration
- Async workflow execution
- Canonical API contracts
- Workflow persistence
- Operational governance
- Architectural invariant enforcement
The system architecture was designed to scale into a real production-grade agricultural intelligence infrastructure.
Challenges we ran into
One major challenge was balancing:
- Investor-grade UI/UX
- Strong architectural engineering
- Limited hackathon resources
- Frontend platform constraints
Another challenge was ensuring architectural discipline while building quickly. We had to carefully separate:
- Workflow orchestration
- Transport adapters
- Business logic
- Operational state management
We also discovered that monitoring very large farms cannot realistically depend only on smartphone uploads, which pushed us toward planning future drone integrations earlier than expected.
Additionally, deployment workflows became difficult because the frontend platform environment lacked terminal access and Git integration support during development.
Accomplishments that we're proud of
We are proud of:
- Designing a cinematic agricultural AI interface
- Creating an operational intelligence experience inspired by enterprise AI systems
- Building scalable orchestration-focused architecture
- Implementing architectural enforcement testing
- Creating AI workflow visibility systems
- Designing the system for future drone-assisted agriculture intelligence
Most importantly, we are proud that AI In Farms was designed as scalable infrastructure — not just a visual prototype.
What we learned
We learned that:
- Strong architecture matters even in hackathons
- AI systems require operational visibility, not just models
- Farmers need low-friction interfaces
- WhatsApp remains a critical access layer in emerging markets
- Architectural enforcement prevents system drift
- Scalability challenges appear earlier than expected in agricultural systems
We also learned how important it is to combine engineering discipline with strong product presentation.
What's next for AI In Farms
Next, we plan to expand AI In Farms into a real agricultural intelligence ecosystem by adding:
- Drone-based crop monitoring
- Real AI disease detection models
- Large-scale farm intelligence scanning
- Predictive agricultural analytics
- Satellite and weather intelligence integration
- Autonomous operational workflows
- Real-time AI orchestration systems
- Multi-region agricultural intelligence infrastructure
Our long-term vision is to build intelligent agricultural infrastructure capable of supporting farmers at scale across Africa and beyond.
Built With
- cnn
- convex
- foundry
- medo
- postgtress
- python
- vs
Log in or sign up for Devpost to join the conversation.