🔑 Pipeline Stages: • Data Injection: Extracted raw data directly from MySQL for structured ingestion • Data Transformation: Cleaned, processed, and feature-engineered datasets for model readiness • Model Training: Automated experiments with MLflow for tracking parameters, metrics, and versions • Data & Model Versioning: Leveraged DVC + DagsHub for reproducible datasets and collaborative workflows • Model Monitoring: Continuous evaluation and logging to ensure reliability and performance Deployment: • Flask API for backend model serving Live Deployed Streamlit App: Hosted interactive dashboard with 50/50 input-output layout, staged animations, and color-coded risk tiers for clarity and trust ⚙️ Tech Stack: GitHub DagsHub MLflow DVC MySQL Flask Streamlit 💡 Outcome: A fully operational MLOps-driven insurance prediction system that makes risk scoring transparent, interactive, and audience-ready. This project demonstrates how modular ML design, reproducible pipelines, and professional dashboards can deliver real-world impact.