You are interviewing for a senior machine learning role and are asked to discuss a past recommendation or prediction project in depth. Use one concrete project as context and answer the following technical decision-making questions: 1. Why did you choose the modeling approach you used? 2. Why did you not use a more complex model? 3. How would you trade off precision, recall, latency, and infrastructure cost? 4. What would you do if offline metrics and online business metrics disagree? 5. How would you tune a decision threshold for a binary classifier? 6. Compare a Factorization Machine with an embedding-based deep neural network for recommendation or ranking.