{"blocks": [{"key": "5ff2a474", "text": "Scenario", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "48eef7db", "text": "Building a real-time fraud-detection system for subscription payments.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "391b5459", "text": "Question", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "a76786e7", "text": "Outline the end-to-end workflow for training and deploying an XGBoost model to flag fraudulent transactions in real time. Which evaluation metrics would you prioritize and why? How would you handle severe class imbalance during training? Describe one strategy for monitoring model drift after deployment.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "285d3b4b", "text": "Hints", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "0d79dc42", "text": "Cover feature engineering, class weighting or sampling, precision-recall trade-offs, and online monitoring.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}], "entityMap": {}}