This project is designed to automate the detection of potentially undervalued iPhones listed on eBay. It integrates eBay’s API with AI-powered item analysis to notify users when good deals appear.
The app uses predefined pricing rules and GPT-based analysis to identify undervalued iPhones listed on eBay. The system operates on a serverless AWS Lambda function that runs every 5 minutes, ensuring real-time monitoring and notifications.
-
Data Collection:
- The app makes periodic API calls to eBay, filtering items based on price, listing time, and specific iPhone models.
- A dictionary of pre-configured models and price ranges is used, allowing for easy scalability.
-
AI-Powered Condition Analysis:
- GPT is integrated to analyze the title and description of each listing. It helps verify the item's condition, ensuring only well-maintained items are flagged for notification.
-
Real-Time Notifications:
- When an undervalued iPhone in good condition is detected, the app sends out real-time notifications to alert the user.
The project delivers notifications via the Pushover service when a new undervalued item is found.
Here’s an example of the notification sent when the app identifies a match:
- Item: Apple iPhone 12 - 64GB - White (Unlocked)
- Price: £150 (which falls within the predefined undervalued price range of £110 - £150)
- Link: The notification includes a direct link to the eBay auction for quick access.
This system ensures that fresh listings from the last 5 minutes are evaluated and flagged in real-time.
- AWS Lambda: Serverless function for executing tasks every 5 minutes.
- eBay API: To search and filter iPhone listings.
- OpenAI (GPT): For analyzing item descriptions to determine item conditions.
- Pushover: For sending real-time push notifications.
- GitHub Actions: Used for CI/CD to automatically update the Lambda function when new code is committed.
The project is set up with a CI/CD pipeline that automatically deploys changes from GitHub to AWS Lambda. The Lambda function is scheduled to run every 5 minutes via a cron job, ensuring up-to-date item listings.
This project showcases my ability to integrate cloud-based services, APIs, and AI models to create real-time automated systems. It demonstrates skills in serverless architecture, automation, and AI-based decision-making.