Crack the ML Recommendation System Interview

Chapter 1: Explicit vs Implicit Signals (/classify-feedback)

Chapter 2: Collaborative Filtering Similarity (/tenant-similar-users)

Chapter 3: Content-Based Filtering & Memory Profiles (/memory-user-profile)

Chapter 4: Hybrid Systems & Weighted Blending (/hybrid-mix-full)

Chapter 5: Knowledge-Based & Cold-Start transitions (/warm-start-sim) + exploration bandit

Chapter 6: Recommender Evaluation Benchmarks (/tenant-evaluate) + measured RMSE/HR@10/NDCG@10

Chapter 7: Neural & Sequential Markov Chain Transitions (/memory-sequence-train) + real two-tower

Chapter 8: Candidate Generation & Approximate Nearest Neighbors (/tenant-scoped-ann) + MMR re-ranking

Chapter 9: Capstone Matrix Factorization Recommender (/capstone-recommend)

Chapter 10: Production Serving & MLOps — caching, retraining/versioning, feature parity, monitoring

Bonus: The ML System Design Casebook — multi-stage funnel, two-tower, multi-task ranking, 20 worked FAANG cases