Automated Student Performance Predictive Analytics Pipeline

Design a comprehensive Python data pipeline using pandas and scikit-learn that ingests multi-year student academic records, preprocesses historical performance data, and builds a machine learning model to predict student at-risk of academic failure. The solution must include feature engineering for academic indicators, handle missing data strategically, implement cross-validation, and generate an interpretable risk score with confidence intervals. Include a Flask-based dashboard that allows educators to input current student data and receive real-time intervention recommendations.