How did we build a smart poultry farming ecosystem in just one year as our Final Year Project?

Our goal wasn't simply to build a dashboard or connect a few sensors to the cloud. We wanted to create a complete platform capable of monitoring poultry farms in real time, analyzing environmental conditions, detecting health issues, and delivering actionable insights to farmers.

The project combines IoT, edge computing, cloud infrastructure, AI, and mobile applications into a single ecosystem.

It starts at the farm

ESP32 associated with dedicated devices continuously collect environmental data such as temperature, humidity, and CO₂ levels.

Instead of sending everything directly to the cloud, the data first passes through an MQTT-based communication layer and a Raspberry Pi edge device responsible for local processing, validation, and synchronization.

This architecture improves reliability and allows the system to continue operating even in challenging farm environments.

Moving to the cloud

Once processed, the data is stored in PostgreSQL hosted on AWS RDS.

On top of that, we built a backend using Django REST Framework, which handles device management, data validation, alert generation, analytics, and dashboard delivery.

The entire backend is containerized with Docker and automatically deployed through GitHub Actions CI/CD pipelines to AWS Elastic Beanstalk.

Real-time monitoring

The telemetry pipeline looks like this:

Despite crossing multiple systems, sensor readings appear on the dashboard almost instantly, giving farmers real-time visibility into farm conditions.

Beyond environmental monitoring

Poultrix goes beyond sensor data.

The platform also integrates:

• ✳ESP32-CAM modules for fecal health monitoring
• ✳CCTV-based computer vision
• ✳AI-powered anomaly detection
• ✳Flutter mobile applications
• ✳Real-time notifications and alerts

What started as an IoT monitoring project evolved into a complete smart farming ecosystem.

In Part 2, we'll break down the AI architecture behind our CCTV analysis, ESP32-CAM disease detection pipeline, and real-time alerting system.