Design User Embedding Semantic Search
Company: Airbnb
Role: Machine Learning Engineer
Category: ML System Design
Difficulty: medium
Interview Round: Onsite
Design a user-embedding-based two-stage semantic retrieval and ranking system for a short-term rental marketplace. The goal is to retrieve and rank property listings that a user is most likely to book, while maximizing booking conversion and long-term user satisfaction.
Address the following:
1. Overall retrieval and ranking architecture.
2. How to build user embeddings.
3. How to build listing embeddings.
4. Training objectives and labels.
5. How to combine semantic relevance with business objectives.
6. How to handle position bias in click and booking logs.
7. Evaluation, experimentation, and monitoring.
Overview: This question evaluates competency in ML system design for recommendation and semantic retrieval, covering user and listing representation learning, two-stage retrieval and ranking architectures, training objectives and label design, bias correction from interaction logs, and evaluation and experimentation practices.
Read the full Airbnb Machine Learning Engineer interview experience this question came from