Airbnb
Online marketplace for lodging with strong data science and infrastructure.
4 Rounds
~21 Days
Hard
The Interview Loop
Recruiter Screen (30 min)
Standard fit check, behavioral questions, and resume overview.
Technical Loop (3-4 Rounds)
Deep dive into domain knowledge, coding, and system design.
Interview Question Bank
Machine Learning Engineer
•
Behavioral
•
easy
Airbnb's core value is 'Be a Host'. Tell me about a time you mentored a junior engineer or went out of your way to help a colleague succeed on a difficult ML project.
#Mentorship
#Collaboration
#Core Values
Machine Learning Engineer
•
Behavioral
•
medium
Tell me about a time your ML model performed well offline but failed or degraded during online A/B testing. How did you debug and resolve the issue?
#Debugging
#Model Deployment
#Offline-to-Online Gap
Machine Learning Engineer
•
Behavioral
•
medium
Tell me about a time you disagreed with a Product Manager regarding the launch of a new ML feature because the model wasn't ready or the metrics were conflicting. How did you navigate the disagreement?
#Conflict Resolution
#Stakeholder Management
#Communication
Machine Learning Engineer
•
Coding
•
medium
Given a list of booking requests represented as intervals [start_date, end_date] and an associated profit for each, write a function to find the maximum profit you can make without accepting overlapping bookings.
#Dynamic Programming
#Sorting
#Intervals
Machine Learning Engineer
•
Coding
•
hard
Given an array of search results represented as strings (host_id, listing_id, score, city), paginate the results such that no two listings from the same host appear on the same page if possible, while maintaining the descending score order. Each page holds exactly N results.
#Hash Maps
#Greedy
#Queues
Machine Learning Engineer
•
Coding
•
medium
Given a dictionary of words and a string of characters (representing a user's misspelled search query), find the longest word in the dictionary that can be formed by deleting some characters from the given string.
#Two Pointers
#String Manipulation
#Sorting
Machine Learning Engineer
•
System Design
•
hard
Design Airbnb's Search Ranking system. How do you balance guest preferences with host acceptance rates?
#Learning to Rank
#Personalization
#Two-sided Marketplaces
Machine Learning Engineer
•
System Design
•
hard
Design the Smart Pricing feature for hosts. What features would you use and how do you handle the cold start problem for a new listing?
#Regression
#Time Series
#Dynamic Pricing
Machine Learning Engineer
•
System Design
•
hard
Design a Trust & Safety ML pipeline to detect fake listings or fraudulent hosts in real-time before they go live on the platform.
#Anomaly Detection
#Classification
#Real-time Systems
#Graph Neural Networks
Machine Learning Engineer
•
System Design
•
medium
Design an LLM-powered system to automatically extract amenities, categorize listing descriptions, and generate a concise summary from unstructured host text and guest reviews.
#Large Language Models
#NLP
#Prompt Engineering
#Data Pipelines
Machine Learning Engineer
•
System Design
•
hard
Design a real-time feature store for Airbnb's ML models. How do you ensure consistency between offline training data and online serving data?
#Feature Store
#Data Engineering
#Stream Processing
#Database Design
Machine Learning Engineer
•
Technical
•
hard
Explain how you would design an A/B test for a new search ranking algorithm. How do you account for network effects and cannibalization in a two-sided marketplace like Airbnb?
#A/B Testing
#Network Effects
#Causal Inference
Machine Learning Engineer
•
Technical
•
medium
In our Learning to Rank (LTR) model for search, we want to optimize for NDCG. Explain the difference between pointwise, pairwise, and listwise approaches. Which would you choose for Airbnb search and why?
#Learning to Rank
#Information Retrieval
#Loss Functions
Machine Learning Engineer
•
Technical
•
hard
How would you handle position bias in Airbnb's search ranking logs when training a new click-through rate (CTR) prediction model?
#Bias Mitigation
#Click Models
#Feature Engineering
Machine Learning Engineer
•
Technical
•
medium
We want to build a model to predict the likelihood of a guest canceling a booking. The dataset is highly imbalanced (cancellations are rare). What metrics would you use to evaluate this model, and how would you handle the class imbalance during training?
#Imbalanced Data
#Evaluation Metrics
#Classification
Difficulty Radar
Based on recent AI-sourced data.
Meet Your Interviewers
The "Standard" Interviewer
Senior EngineerFocuses on core competencies, system constraints, and clear communication.
SimulateUnwritten Rules
Think Out Loud
Always explain your thought process before writing code or drawing architecture.