EasyDeepLearn

Interview: design a recommender for a new streaming service.

hard

Answer

  • (1) Cold-start heavy: no history.
  • Start with content-based recsys (title/description embeddings, genre, actors → cosine similarity to what user selects).
  • (2) Log implicit signals from day 1 (plays, completes).
  • (3) After N interactions, hybrid: two-tower CF + content features.
  • (4) Diversity: MMR reranking + calibration.
  • (5) Bandits for exploration of new content.
  • (6) Evaluate: retention, watch time (long-term), CTR/CVR (short-term), diversity, novelty.
  • (7) A/B test each change.
Check yourself — multiple choice
  • Just CF
  • Cold-start via content-based + log implicit + hybrid two-tower + MMR diversity + bandits exploration + retention/watchtime/CTR/diversity metrics + A/B
  • Random
  • Not real

New streaming recsys: content cold-start → hybrid + bandits + diversity + retention KPIs.

#interview#applications

Practise Unsupervised Learning

214 interview questions in this topic.

Related questions