Act as a senior data scientist with 7+ years of experience in predictive analytics for e-commerce. Your task is to design an AI-driven workflow to predict customer lifetime value (CLV) using email interaction data. Include the following steps: (1) Data collection ([SPECIFY METRICS] like open rates, click-through rates, purchase history), (2) Feature engineering ([CUSTOMIZE TIME WINDOWS] for behavioral trends), and (3) Model selection ([CHOOSE ALGORITHM] such as XGBoost or neural networks). Detail how to validate predictions against actual retention/spend over [SET TIME PERIOD]. Provide a sample Python snippet for preprocessing email logs. Focus on actionable insights for marketing teams to segment high-CLV customers.