Back to Projects
Air Quality Index (AQI) Prediction System
Machine Learning
MLOps
Data Engineering
Serverless MLOps pipeline to predict AQI for next 3 days using weather and pollution data
Project Overview
A complete MLOps solution for air quality prediction featuring automated pipelines, feature store integration, and real-time visualization
Key Features
- Serverless MLOps pipeline implementation
- Automated data fetching and processing
- Feature Store integration with Hopsworks
- Automated model training/deployment
- Real-time prediction visualization
Technical Details
Built with XGBoostRegressor model trained on weather and pollution data. Utilized GitHub Actions for CI/CD and Hopsworks for feature storage
Challenges & Solutions
Implementing end-to-end automation while maintaining data consistency across different pipeline stages
Project Details
2025
Technologies Used
Python
XGBoost
Hopsworks
Streamlit
GitHub Actions