Design a dynamic rental pricing system
Overview
This question evaluates skills in production machine learning system design for pricing, including problem formulation, data and feature engineering, modeling and uncertainty management, online inference and scalability, experimentation and guardrails, and considerations like fairness and cold-start handling.
Airbnb
Aug 12, 2025, 12:00 AM
Software Engineer
Onsite
ML System Design
0
0
System Design: ML-Driven Nightly Pricing for Short-Term Rentals
Context
Design a production ML system that recommends (and optionally auto-sets) nightly prices for hosts on a two-sided rentals marketplace. The system should maximize long-term marketplace health while protecting hosts and guests with business guardrails.
Requirements
-
Problem formulation
- Objective(s)
- Constraints and business guardrails
-
Data sources
- Historical bookings, search demand, competitor prices, calendars, local events
-
Feature engineering
- Seasonality, lead time, availability/inventory, price elasticity, cancellations
-
Modeling approach
- E.g., time-series + gradient boosting with elasticity estimation, or constrained reinforcement learning
- How to incorporate uncertainty and guardrails
-
Training pipeline and evaluation
- Offline training, offline simulation/sandboxing
-
Online inference and architecture
- Service design, latency and scale targets
- Exploration–exploitation strategy
- Handling cold-start listings and sparse regions
- Fairness, explainability, and abuse prevention
- Rollout plan with A/B testing and guardrail metrics
Solution
Login required
Show