The diagram shows the system architecture
The call log from Twilio showed the activity when I recorded the demo video.
The AWS services that were used.
The monitor dashboard showed when the AWS Lambda function was called.
The CloudWatch log of AWS Lambda when Lambda is triggered

WhiskerWatch: Smart Cat Food Monitoring System

Inspiration

As a busy cat parent, I often found myself forgetting to refill my cat's food container until it was completely empty. This would lead to last-minute store runs or, worse, a very unhappy feline friend. I realized that technology could help solve this common problem that many pet owners face, leading to the creation of WhiskerWatch.

What it does

WhiskerWatch is an automated cat food monitoring system that take advantage of event-driven serverless architecture:

Captures daily images of the cat food container using a repurposed laptop camera
Automatically uploads these images to an AWS S3 bucket
Processes the images using AWS Lambda and Amazon Rekognition to determine food levels
Triggers a phone call via Twilio API when food levels are critically low
All infrastructure is managed through Infrastructure as Code using Terraform

How we built it

The system architecture consists of several key components:

Python script running on a laptop to capture and upload images
Storage: AWS S3 bucket for image storage
Processing: AWS Lambda function triggered by S3 uploads
Analysis: AWS Rekognition for image processing and food level detection
Notification: Twilio API integration for automated phone calls
Infrastructure: Terraform configurations for AWS resource provisioning

Challenges we ran into

The main technical challenges included:

Debugging Lambda functions in the cloud environment proved more complex than local development
Managing AWS IAM permissions and roles through Terraform