21. Recommend the right items to the right people
Use user-item data, collaborative filtering, ranking metrics, candidate generation, and re-ranking. You will design recommendation systems that balance relevance, freshness, diversity, and business constraints.