Interview questions
Common questions for roles at Meta.
17 questions for 3 roles: Backend engineer, Mobile engineer, AI/ML engineer.
Free with every account.
Practice questions from open collections and written with AI. Not reported from interviews.
What shaped them: what Meta publicly builds, social apps (Facebook, Instagram, WhatsApp) and their feeds and messaging.
By type: technical 15, behavioral 2. Of them, coding 2.
Backend engineer
6 questions
Technical
Explain how you would implement message queueing to handle millions of concurrent chat messages ensuring delivery and ordering.
Given a large volume of user-generated content, how would you efficiently detect and filter out duplicate posts or spam?
Describe how you would optimize database schema and queries for storing and retrieving social graph connections in a social app.
Can you discuss the challenges of ensuring data consistency and availability in a distributed messaging system?
What privacy and security considerations are critical when building social apps that handle personal messages and feeds?
Behavioral
Tell me about a time you had to troubleshoot a complex backend performance issue under pressure and how you resolved it.
Mobile engineer
5 questions
Technical
How would you implement infinite scrolling in a social media feed while minimizing memory usage and network calls on Android or iOS?
Describe an approach to handle real-time messaging delivery and read receipts in a chat application with unreliable network conditions.
Can you write a function that merges and sorts posts from multiple user feeds into one chronological timeline efficiently?
Coding
What strategies would you use to optimize image loading and caching in a mobile social app to improve performance and user experience?
How do privacy and data security considerations affect your design of features like message encryption or content sharing in social apps?
AI/ML engineer
6 questions
Technical
How would you design a feed ranking system that balances content personalization with freshness for millions of users simultaneously?
Explain how you would detect and filter out spam or abusive messages in a large-scale messaging platform using machine learning.
Given a skewed dataset of user interactions with posts, what techniques would you use to build an effective click-through rate prediction model?
Write code to preprocess and tokenize text messages for input into a recurrent neural network, considering different languages and emojis.
Coding
How would you handle cold-start problems for new users in a social app's recommendation system?
Behavioral
Describe a time when you improved the performance or scalability of a machine learning model under production constraints.
Next
When the interview comes.
Transcribe
Live transcription of a call, readable on any phone as it happens.
Learn moreInterview copilot
Suggested answers during your interview, from your resume and the role.
Learn more