Capstone: An Automated Retraining Pipeline

Day 70 of 100 Days of MLOps, and the Module 7 finale. Assemble everything from the module into one real automated retraining pipeline: a Prefect flow that ingests, validates, trains, and evaluates behind a quality gate — parameterized, retried, schedulable nightly, observable in the UI, and wired to alert on failure. You'll run it end-to-end (a healthy retrain that deploys, and a gate failure that blocks a bad model and fires an alert). It's the self-running retraining loop the whole module was building toward. Runs 100% locally on any OS.