← Airbnb Interview Insights

Airbnb·Machine Learning Engineer·Onsite - System Design / Architecture·Senior

SeniorPrefer not to say
Jul 2026

Summary

ML system design round at Airbnb for an MLE role. The whole thing was a single deep-dive on building a listing LTV model, and it went way longer and wider than I expected. There's a lot of surface area here.

Questions Asked (8)

Q1

Design an ML system to predict the lifetime value of an Airbnb listing. Walk through your end-to-end approach.

System DesignTechnical Trade-offsProduct Analytics & Metrics
Author's notes

This is one of those questions where the scope keeps expanding the more you talk.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the business definition of LTV for Airbnb listings, then outline an end-to-end ML pipeline covering data collection, feature engineering, model selection, and deployment. Emphasize trade-offs between model complexity, interpretability, and business impact, and discuss how to handle challenges like seasonality and sparse data.

Pro tip: Frame LTV as a forward-looking metric and propose using survival analysis or probabilistic models to handle censored data, showing depth beyond standard regression. Also, tie model outputs to actionable business decisions like pricing or marketing spend to demonstrate product sense.

1. Clarify Business Objective and Define LTV

Ask clarifying questions to understand how Airbnb defines LTV for listings (e.g., total revenue over a horizon, net of costs) and the prediction horizon. Align on the target variable and evaluation metrics.

2. Data Collection and Feature Engineering

Identify data sources: booking history, listing attributes, host behavior, seasonality, market trends, and external factors. Engineer features like booking frequency, average daily rate, occupancy rate, host responsiveness, and temporal patterns.

3. Model Selection and Training

Choose appropriate models: start with regression (e.g., linear, tree-based) but consider survival analysis (e.g., Cox proportional hazards) or probabilistic models to handle censored data. Address challenges like seasonality, sparsity, and non-stationarity.

4. Evaluation and Validation

Use time-based cross-validation to avoid leakage. Evaluate with metrics like MAE, RMSE, and business metrics (e.g., revenue lift from targeting). Consider calibration and uncertainty quantification.

5. Deployment and Monitoring

Deploy model as a service for batch or real-time predictions. Monitor performance, data drift, and business impact. Set up retraining pipelines and feedback loops.

Key Points to Mention

  • Definition of LTV: revenue-based vs. profit-based, time horizon, and how it aligns with business goals.
  • Handling censored data: many listings are still active, so survival analysis or probabilistic models are more appropriate than standard regression.
  • Feature engineering: incorporate temporal dynamics (seasonality, trends), host behavior, and market competitiveness.
  • Model choice trade-offs: interpretability (e.g., linear models) vs. predictive power (e.g., gradient boosting), and the need for uncertainty estimates.
  • Evaluation: use time-based splits, business metrics, and consider long-term impact vs. short-term accuracy.
  • Deployment: scalability, latency, monitoring for drift, and integration with business systems (e.g., pricing, marketing).

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

How would you define LTV for this problem, and what clarifying questions would you ask before modeling anything?

Product Sense & IdeationAdaptability & Ambiguity
Author's notes

Actually felt okay on this one.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining LTV as a forward-looking, discounted sum of net contribution margin per user over a chosen horizon, then immediately pivot to clarifying questions that pin down the business objective, user unit, and measurement constraints. Show that you tailor the definition to the specific problem (e.g., guest vs. host LTV, booking vs. revenue) before jumping into modeling.

Pro tip: Anchor your clarifying questions around the decision the model will inform—this signals product sense and prevents you from over-engineering a metric that doesn't drive action. Also, mention that LTV should be net of variable costs and discounted, not just gross revenue, to show financial maturity.

1. Define LTV in context

State a working definition: LTV is the expected net profit (or contribution margin) attributable to a user over a specified time horizon, discounted to present value. Clarify whether we mean guest, host, or both, and whether it's per booking, per trip, or per user.

2. Ask clarifying questions

Ask about the business goal (e.g., acquisition, retention, pricing), the unit of analysis (user, listing, market), the time horizon, and what costs to include. Also ask about data availability, seasonality, and whether we need individual-level or cohort-level LTV.

3. Identify modeling constraints

Discuss data limitations (censored data, long tails, multi-sided marketplace), the need for causal inference vs. prediction, and how the model will be evaluated (e.g., calibration, lift in downstream decisions).

4. Propose a modeling approach

Outline a high-level plan: e.g., use probabilistic models (BG/NBD, Gamma-Gamma) for repeat purchases, incorporate survival analysis for churn, and adjust for marketplace dynamics. Emphasize starting simple and iterating.

5. Validate and iterate

Describe how you would validate the LTV model (holdout, backtesting, sensitivity analysis) and how you would monitor it in production, including feedback loops and potential biases.

Key Points to Mention

  • LTV should be net of variable costs (e.g., host payouts, payment processing, customer support) and discounted to present value.
  • Clarify the time horizon: short-term (e.g., 1 year) vs. long-term (lifetime) and how that affects modeling choices.
  • Distinguish between guest LTV and host LTV, as they have different drivers and retention dynamics on Airbnb.
  • Consider marketplace effects: a guest's LTV may depend on host quality and vice versa, requiring a multi-sided approach.
  • Address data challenges: censored data (users still active), seasonality, and the need for survival analysis or probabilistic models.
  • Tie the LTV definition to the specific decision it will inform (e.g., marketing spend, product changes) to ensure relevance.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

What data sources and features would you use, and how would you engineer them for this model?

Data ModelingSystem Design
Author's notes

Covered the obvious stuff: historical bookings, listing attributes, host history, market-level demand signals, seasonality, reviews, photos.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the model's objective and business context, then outline the data sources (e.g., user behavior logs, listing details, transaction records) and how you would transform them into features. Emphasize a systematic feature engineering pipeline that handles temporal aspects, categorical variables, and scalability, while ensuring data quality and avoiding leakage.

Pro tip: Demonstrate awareness of Airbnb's unique data challenges, such as sparse user-item interactions and the importance of location-based features, and mention how you would validate feature importance and monitor drift in production.

1. Clarify the Model Objective and Business Context

Ask questions to understand what the model predicts (e.g., booking likelihood, price suggestion) and how it will be used, as this determines relevant data sources and features.

2. Identify and Evaluate Data Sources

List potential data sources such as user activity logs, listing attributes, search queries, reviews, and external data (e.g., local events), and discuss their availability, reliability, and ethical considerations.

3. Design Feature Engineering Pipeline

Describe how to transform raw data into features: aggregations (e.g., user booking history), temporal features (e.g., time since last booking), categorical encoding (e.g., target encoding for neighborhoods), and text embeddings for reviews.

4. Address Scalability and Productionization

Explain how to implement the pipeline efficiently (e.g., using Spark, Airflow) and ensure features are computed consistently in training and serving, with monitoring for drift and freshness.

5. Validate and Iterate on Features

Discuss methods to evaluate feature importance (e.g., SHAP, permutation importance) and iterate based on model performance and business metrics.

Key Points to Mention

  • Temporal features and time-based splits to prevent data leakage
  • Handling categorical variables with high cardinality (e.g., listing IDs, neighborhoods) using target encoding or embeddings
  • Aggregation features from user and listing history (e.g., average rating, booking frequency)
  • Text and image features from reviews and listing photos using NLP and computer vision
  • Scalable data processing with distributed computing (e.g., Spark) and feature stores
  • Monitoring feature drift and ensuring reproducibility in production

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q4

Many listings are still active and haven't reached the end of their lifetime. How do you handle censored data in your training set?

Technical Trade-offsAlgorithms & Data Structures
Author's notes

Blanked for a second.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by acknowledging that censored data is common in time-to-event problems like booking or churn, and that ignoring it biases the model. Then explain how you would use survival analysis techniques such as Cox proportional hazards or accelerated failure time models, which naturally handle right-censoring. Finally, discuss practical implementation details like feature engineering, evaluation metrics, and trade-offs with simpler approaches.

Pro tip: Mention that you would validate the proportional hazards assumption and consider time-varying covariates if the effect changes over time, showing depth beyond textbook survival analysis.

1. Define the problem and censoring type

Clarify that the goal is to model time-to-event (e.g., time until booking or listing removal) and that right-censoring occurs when the event hasn't happened by the observation window. Confirm the censoring mechanism is independent of the event time.

2. Choose a survival analysis model

Select an appropriate model such as Cox proportional hazards, accelerated failure time, or discrete-time survival models. Discuss how these models incorporate censored observations into the likelihood function without discarding them.

3. Engineer features and handle time-varying covariates

Create features that capture listing characteristics, host behavior, and temporal patterns. If covariates change over time (e.g., price updates), use extended Cox models or time-varying covariates to avoid bias.

4. Evaluate with appropriate metrics

Use concordance index (C-index), time-dependent AUC, or Brier score to assess predictive performance. Avoid using standard regression metrics like RMSE, which ignore censoring.

5. Discuss trade-offs and alternatives

Compare survival analysis with simpler approaches like treating censored as non-events (biased) or imputing event times (risky). Highlight computational complexity and interpretability trade-offs.

Key Points to Mention

  • Right-censoring and its impact on bias if ignored
  • Cox proportional hazards model and partial likelihood
  • Accelerated failure time models and parametric assumptions
  • Concordance index (C-index) for evaluation
  • Time-varying covariates and proportional hazards assumption
  • Trade-offs between survival analysis and simpler classification/regression approaches

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q5

Would you consider a two-stage modeling approach here? What would that look like?

System DesignTechnical Trade-offs
Author's notes

They basically handed me the answer by asking the question this way.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Acknowledge that a two-stage approach is often valuable for complex problems like ranking or recommendation, then outline the stages (e.g., candidate generation and ranking) and explain the rationale, trade-offs, and implementation details. Emphasize how this architecture improves scalability, accuracy, and maintainability.

Pro tip: At Airbnb, two-stage modeling is common in search and recommendation; mention how you would monitor and iterate on each stage separately, and how you'd handle feedback loops between them.

1. Clarify the problem and goals

Restate the problem to ensure alignment, and identify key objectives such as latency, accuracy, and scalability that motivate a two-stage approach.

2. Define the stages

Describe the two stages: typically a lightweight candidate generation model (e.g., retrieval) followed by a more complex ranking model. Explain the role of each.

3. Explain the benefits and trade-offs

Discuss advantages like reduced computational load, ability to use complex models on a smaller set, and improved relevance. Acknowledge trade-offs such as potential recall loss and increased system complexity.

4. Outline implementation details

Cover how you would train, evaluate, and deploy each stage, including data flow, feature consistency, and how to handle cold start or new items.

5. Discuss evaluation and iteration

Explain how you would measure end-to-end performance and each stage's contribution, and how you would iterate to improve both stages over time.

Key Points to Mention

  • Candidate generation vs. ranking: different model complexities and objectives
  • Scalability: handling large item catalogs efficiently
  • Latency and throughput considerations in production
  • Trade-offs: recall vs. precision, complexity vs. maintainability
  • Evaluation metrics: offline (e.g., NDCG, recall@k) and online (e.g., CTR, bookings)
  • Airbnb-specific context: search ranking, recommendations, and personalization

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q6

How do you handle the cold-start problem for brand new listings with no booking history?

Technical Trade-offsProduct Sense & Ideation
Author's notes

Pretty standard cold-start framing.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the cold-start problem as a data scarcity issue and outline a multi-pronged strategy: leveraging content-based features, transfer learning from similar listings, and exploration-exploitation techniques. Emphasize the need to balance short-term performance with long-term data collection, and discuss how to evaluate and iterate on the solution.

Pro tip: Show awareness of the business context: new listings are critical for supply growth, so your solution should prioritize giving them a fair chance while protecting the user experience. Mention the importance of logging and feedback loops to continuously improve the model.

1. Define the problem and constraints

Clarify what 'cold-start' means for Airbnb listings: no booking history, but possibly rich metadata (photos, amenities, host info, location). Discuss constraints like user experience, fairness, and scalability.

2. Leverage content and context features

Use listing attributes (e.g., images, text descriptions, amenities, location) and host features to build an initial ranking model. Mention techniques like embeddings, computer vision for photos, and NLP for descriptions.

3. Transfer learning and meta-learning

Exploit data from similar listings or hosts to make predictions. For example, use a hierarchical model or meta-learning to learn from listings in the same city or category, or use host-level history if available.

4. Exploration and active learning

Implement a bandit-based approach to allocate some traffic to new listings to gather feedback. Balance exploration (showing new listings) with exploitation (showing proven listings) using techniques like Thompson sampling or epsilon-greedy.

5. Evaluation and iteration

Define metrics (e.g., booking rate, user satisfaction) and set up A/B tests to measure the impact. Use offline evaluation with simulated data and online experiments to refine the model over time.

Key Points to Mention

  • Content-based filtering using listing features (images, text, amenities)
  • Transfer learning from similar listings or hosts
  • Exploration-exploitation trade-off (multi-armed bandits)
  • Use of auxiliary data (host history, location trends)
  • Evaluation metrics and A/B testing for cold-start models
  • Business impact: supporting new hosts and supply growth

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q7

What offline and online metrics would you use to evaluate this system, and how do you connect them to business outcomes?

A/B Testing & ExperimentationProduct Analytics & MetricsTechnical Trade-offs
Author's notes

Went through RMSE and calibration for the regression case, top-K precision for the bucketed classification version.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the system's purpose and business goals, then define offline metrics for model performance and online metrics for user and business impact. Explain how offline metrics inform online experiments and how online results tie to key business outcomes like bookings and revenue.

Pro tip: Emphasize that offline metrics are proxies and must be validated through online experiments; always connect metrics to a north-star business metric and consider guardrail metrics to avoid negative side effects.

1. Clarify system and business objectives

Identify what the system does (e.g., search ranking, recommendation) and the key business goals it supports, such as increasing bookings or guest satisfaction.

2. Define offline metrics

Select metrics that evaluate model performance on historical data, such as precision, recall, NDCG, or RMSE, depending on the task.

3. Define online metrics

Choose metrics that measure real-world impact through A/B tests, such as click-through rate, conversion rate, and booking value.

4. Connect offline to online

Explain how offline improvements are expected to translate to online gains, and how you validate this through experimentation.

5. Link to business outcomes

Map online metrics to ultimate business KPIs like revenue, bookings, or retention, and discuss trade-offs and guardrail metrics.

Key Points to Mention

  • Offline metrics: precision, recall, NDCG, AUC, RMSE, etc., depending on model type.
  • Online metrics: CTR, conversion rate, booking rate, revenue per user, session duration.
  • A/B testing framework and statistical significance.
  • North-star metric and guardrail metrics to monitor health.
  • Long-term effects and counterfactual evaluation.
  • Business outcomes: bookings, revenue, customer retention, host satisfaction.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q8

How would you deploy this model and how often would you retrain or refresh it?

System Design
Author's notes

Talked through batch vs near-real-time scoring, storing predictions in a feature store, and triggering retraining on data drift or a fixed cadence.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the model's purpose and requirements (e.g., real-time vs batch, latency, scale). Then outline a deployment architecture that fits Airbnb's infrastructure, covering serving, monitoring, and CI/CD. Finally, discuss retraining strategy based on data drift, performance metrics, and business cadence, emphasizing automation and experimentation.

Pro tip: Tie your answer to Airbnb's specific context—mention their ML platform (Bighead, Zipline, etc.) and how models like search ranking or pricing are deployed. Show awareness of trade-offs between freshness and cost, and propose a hybrid approach (e.g., online learning for some features, periodic retraining for others).

1. Clarify requirements and constraints

Ask about the model's use case, expected traffic, latency requirements, and data availability. This ensures your deployment and retraining plan is tailored to the problem.

2. Design deployment architecture

Propose a serving infrastructure (e.g., real-time API, batch prediction, or streaming) using tools like Kubernetes, TF Serving, or Airbnb's ML platform. Include model versioning, canary rollout, and A/B testing.

3. Implement monitoring and observability

Describe how you'll track model performance (accuracy, latency, drift) and system health (resource usage, errors). Set up alerts and dashboards for proactive maintenance.

4. Define retraining/refresh strategy

Explain triggers for retraining (e.g., time-based, drift detection, performance degradation) and the retraining pipeline (data collection, training, validation, deployment). Discuss online learning if applicable.

5. Address trade-offs and iteration

Acknowledge trade-offs between retraining frequency and cost/complexity. Propose a feedback loop for continuous improvement and experimentation.

Key Points to Mention

  • Model serving options: real-time (REST/gRPC), batch, or streaming, with considerations for latency and throughput.
  • Use of feature stores (e.g., Airbnb's Zipline) to ensure consistency between training and serving.
  • Canary deployment and A/B testing to safely roll out new model versions.
  • Monitoring for data drift, concept drift, and model performance degradation.
  • Retraining triggers: scheduled (e.g., daily/weekly), performance-based, or drift-based.
  • Automated retraining pipelines (CI/CD/CT) and version control for reproducibility.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.