01 / Overview
Client platform · full-stack delivery
An IoT platform for smart agriculture
KMC Agri is an IoT platform owned by CÔNG TY TNHH MTV TV TM DV KIM MÃ and developed to manage farms and greenhouses.
Roam Moon built the complete system, including:
- Web management dashboard built with Next.js
- Backend built with NestJS
- Android and iOS applications
Operators can monitor sensors, control devices, configure automation, review activity history, receive notifications, and manage every area from one unified platform.
02 / Challenge
A farm may have many sensors and devices operating at the same time.
The system needed to:
- Collect data continuously
- Control devices remotely
- Configure automation
- Synchronize across Web, Android, and iOS
- Remain stable even when network quality is poor
The platform also needed to stay simple for operators while remaining flexible enough to support future expansion.
03 / Goals
- Monitor the entire farm in real time
- Control devices from anywhere
- Automate care processes
- Detect issues early and send alerts
- Build a platform that can scale across multiple farms
04 / Technical problems
Standardizing many device types
Each sensor and controller may use a different protocol, but all of them must operate through one consistent system.
Synchronizing data in real time
Changes made on the Web or a phone must appear immediately across every connected client.
Ensuring dependable control commands
Device on/off commands must be processed accurately even when the network is intermittent.
05 / Research
The team studied real operating workflows at the farm, covering:
- Devices
- Sensors
- Irrigation systems
- Greenhouses
- Care processes
- Alerts
- Automation
We then analyzed user needs to design an architecture that serves current operations and can continue expanding in the future.
06 / Architecture
The system’s core components include:
- Next.js Web dashboard
- NestJS backend
The backend manages:
- Farms
- Devices
- Sensors
- Automation
- History
- Notifications
- Reports
- Device connectivity
- Realtime services
Data is stored in MongoDB and synchronized instantly to Web and mobile clients through WebSocket.
The architecture is designed to be:
- Secure
- Easy to scale
- Easy to monitor
- Independently operating components
Reference architecture
Safe and stable operations
Devices & Edge
Sensors · PLC · Gateway
Connectivity
MQTT · HTTP · gRPC
Orchestration
Message Queue · Event Bus · Automation
Services
Device Management · API · Notification
Storage
Time-series · MongoDB · Object Storage
Applications
Web · Mobile
07 / Design
The interface is optimized for both operations engineers and everyday users.
The most important information is always presented clearly:
- Sensor status
- Active devices
- Alerts
- Required actions
A consistent experience across Web and mobile helps users move between devices without having to relearn the system.
Interface gallery
Explore the KMC Agri operations experience across Mobile and Web. Select an image to view it at high resolution.
6 images
Mobile application
04 screensWeb dashboard
02 screens08 / Development
The system is divided into independent modules, including:
- Devices
- Sensors
- Automation
- Notifications
- History
- Reports
- Finance
Each capability is developed, tested, and deployed independently to reduce the risk of upgrades.
09 / Optimization
- Realtime synchronization removes the need for constant manual refreshes
- Device Profiles make new hardware faster to add without disrupting the existing system
10 / Deployment
The system is released in controlled stages to reduce operational risk. Every update is:
- Monitored in operation
- Checked for errors
- Prepared for rollback when necessary
11 / Scaling
The architecture is designed to:
- Add more farms
- Add more devices
- Add more users
- Expand individual modules without affecting the entire system
This approach keeps the platform stable as its operating scale grows.
12 / Monitoring
The system continuously monitors:
- Sensor data
- Device status
- Realtime services
- Push notifications
- Application versions
- Activity logs
This visibility helps the operations team detect and resolve incidents faster.
13 / Result
One unified operations center
Monitor complete farm status across Web and Mobile.
Process automation
Reduce manual work through sensor-data-driven control scenarios.
Faster incident response
Alerts and device states update in real time.
Ready for long-term growth
Easily add sensors, devices, and operational areas.
14 / Lessons learned
The project demonstrates that a sustainable IoT system depends not only on devices or sensors, but also on how its software architecture is designed.
When components are clearly separated, data is standardized, and operating processes are unified, the system becomes easier to scale, maintain, and adapt to new requirements in the future.
Our responsibilities
Duration
6 months
Deployment date