RentalAI is a rental price prediction platform built with a modern microservices architecture.
It combines real estate data, statistical modeling, and interactive visualizations to help users explore rental markets and predict fair rent prices.
Live site: https://rentalai.vercel.app
Housing affordability and fair rent pricing are difficult for renters and landlords to estimate. RentalAI addresses this by providing:
- Interactive property listings with search and mapping
- Rental price predictions based on historical data
- Visual analytics for rent distribution and trends
- A clear, documented architecture for developers
RentalAI uses a microservices architecture where each component has a specific responsibility and can be scaled independently.
The frontend communicates with both the main backend API and a Random Forest ML service for predictions. Data visualizations are stored in AWS S3 for efficient delivery, while MongoDB stores test and property data for the ML service.
- Realtor API (property listings)
- OpenStreetMap (geolocation & mapping)
These APIs are wrapped and used for initial data seeding.
- Next.js
- React
- TypeScript
- Tailwind CSS
- Leaflet (maps)
- Flask
- Python
- Pandas
- Swagger (API documentation)
- CORS
- Flask
- Scikit-learn (Random Forest)
- NumPy
- Pandas
- Pickle (model serialization)
- Jupyter Notebook
- MongoDB (NoSQL)
- AWS S3 (images & ml data charts)
| Endpoint | Description |
|---|---|
GET /api/get_data |
Retrieve property listings |
GET /api/get_rent_by_month |
Returns rent-over-time chart |
GET /api/get_rent_distr |
Returns rent distribution |
GET /api/get_image_paths |
Image paths for listings |
| Endpoint | Description |
|---|---|
POST /api/get_prediction |
Predict rental price from features |
GET /api/get_importance |
Feature importance visualization |
GET /api/get_test_data |
Retrieve test dataset |
Frontend: RentalAI/front-end
Backend API: RentalAI/backend-api-service
ML Service: RentalAI/ml-service
MIT © 2026 RentalAI — By Yassine Moumine
