{"blocks": [{"key": "d40db2a3", "text": "Scenario", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "1c25f6f0", "text": "You own a churn-prediction pipeline that trains weekly on 10M users.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "1f9b2711", "text": "Question", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "30d51353", "text": "Walk me through feature engineering, model selection and hyper-parameter tuning for churn prediction. Why might you favor Gradient Boosted Trees over Logistic Regression here? Describe two techniques for explaining model outputs to non-technical stakeholders. If recall suddenly drops by 15% week-over-week, outline a debugging checklist.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "af8c697d", "text": "Hints", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "3f15a9ea", "text": "Discuss imbalance handling, SHAP, feature drift, and offline/online parity checks.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}], "entityMap": {}}