Facebook Interview Questions and Experience

1. What is the difference between concurrency and parallelism?
   Concurrency is handling multiple tasks at once (not necessarily simultaneously), while parallelism is executing tasks simultaneously using multiple processors.
2. Explain the difference between TCP and UDP.
   TCP is connection-oriented, reliable, and ensures data delivery. UDP is connectionless, faster, but doesn’t guarantee delivery or order.
3. What is a race condition, and how do you prevent it?
   A race condition occurs when multiple threads access shared data simultaneously. It is prevented using locks, semaphores, or atomic operations.
4. Explain how a hash table works.
   A hash table maps keys to values using a hash function. Collisions are handled with techniques like chaining or open addressing.
5. What are the differences between process and thread?
   A process is an independent execution unit with its own memory, while threads share memory within a process, making them lightweight.
6. How does garbage collection work in Java?
   Garbage collection automatically removes unused objects to free memory, using algorithms like Mark-and-Sweep or Generational Collection.
7. What is the difference between SQL and NoSQL?
   SQL is structured, relational, and schema-based. NoSQL supports unstructured data, high scalability, and flexibility for large-scale applications.
8. What is sharding in databases?
   Sharding is horizontal partitioning of data across multiple servers to improve performance and scalability in large distributed databases.
9. Explain CAP theorem in distributed systems.
   CAP theorem states a distributed system can only guarantee two of Consistency, Availability, and Partition Tolerance simultaneously.
10. What is CDN and why is it important?
    A Content Delivery Network caches content across servers worldwide, reducing latency and improving website performance for global users.
11. What is a deadlock, and how can it be avoided?
    Deadlock happens when processes wait indefinitely for resources. Avoid it using ordering, timeout, or avoiding circular wait.
12. What are load balancers and their types?
    Load balancers distribute traffic across servers. Types include Layer 4 (transport-level) and Layer 7 (application-level) balancing.
13. Explain the difference between REST and GraphQL.
    REST uses fixed endpoints and returns full data, while GraphQL allows flexible queries, fetching only required fields.
14. What is an API rate limiter, and why is it needed?
    A rate limiter restricts the number of API calls in a timeframe, preventing abuse and maintaining system stability.
15. What is idempotency in APIs?
    Idempotency ensures multiple identical requests have the same effect, preventing issues like duplicate payments.

For Freshers (SDE-1)

1. Why do you want to join Meta as a fresher?
   Meta offers challenging problems at scale and opportunities to learn from world-class engineers, making it ideal for career growth.
2. How do you approach solving algorithmic problems?
   I start with brute force, identify inefficiencies, optimize step by step, and clearly explain my thought process to the interviewer.
3. What’s your favorite project, and why?
   My favorite project was a social media clone app because it taught me about scalability, authentication, and real-world system design basics.
4. How do you keep improving your problem-solving skills?
   By solving DSA problems daily, participating in coding competitions, and reviewing solutions to learn alternative approaches.

For Data Analysts

1. How would you analyze user engagement on Facebook posts?
   By tracking KPIs like likes, shares, comments, time spent, and using visualization tools for trend analysis.
2. How do you deal with missing data in a dataset?
   I use techniques like mean/median imputation, forward fill, or removing incomplete rows depending on context.
3. Explain the difference between supervised and unsupervised learning.
   Supervised learning uses labeled data for prediction, while unsupervised learning finds hidden patterns in unlabeled data.
4. How would you measure the success of a new feature in Facebook?
   By defining KPIs like daily active users, retention, and engagement rate, then running A/B testing.

For Backend Developers

1. How would you design a scalable messaging system like Messenger?
   By using distributed databases, message queues, load balancers, and ensuring fault tolerance with replication.
2. Explain the role of caching in backend systems.
   Caching reduces latency and database load by storing frequently accessed data closer to the application.
3. How would you handle billions of API requests daily?
   Using load balancers, sharding, asynchronous processing, and distributed queues to ensure system scalability.
4. What’s the difference between synchronous and asynchronous communication in services?
   Synchronous waits for response (blocking), while asynchronous allows tasks to continue without waiting, improving throughput.

For Managers

1. How do you prioritize tasks when leading multiple projects?
   I balance urgency, impact, and resources, using frameworks like MoSCoW or RICE for prioritization.
2. How do you resolve conflicts in your team?
   By facilitating open communication, understanding both perspectives, and reaching a solution aligned with project goals.
3. How do you measure success of your team?
   Through delivery timelines, code quality, project outcomes, and peer reviews, rather than just hours spent.
4. How do you motivate engineers during challenging projects?
   By acknowledging difficulties, setting achievable goals, and recognizing individual contributions to maintain morale.

1. Tell me about a time you had to work with a difficult teammate.
   I stayed professional, focused on shared goals, and resolved differences through open communication.
2. Describe a situation where you had to learn a new skill quickly.
   During an internship, I learned Docker within a week by following tutorials and applying it to project deployment.
3. How do you handle failure in projects?
   I analyze what went wrong, learn from mistakes, and implement corrective measures for future success.
4. Tell me about a time when you showed leadership in a project.
   I delegated tasks, maintained regular check-ins, and ensured deadlines were met, helping the team succeed.
5. How do you handle high-pressure deadlines?
   By breaking tasks into smaller milestones, prioritizing effectively, and staying calm under pressure.
6. Describe a time when you gave constructive feedback.
   I pointed out inefficiencies in a peer’s code, suggested improvements, and helped implement them together.
7. How do you adapt to changing project requirements?
   I reassess priorities, adjust timelines, and collaborate with stakeholders to align with new goals.
8. Give an example of when you worked across teams.
   In a group project, I collaborated with design and testing teams, ensuring smooth delivery through active communication.

Candidate Background (Fresher, skills, education):

3rd-year B.Tech (CSE) at JIIT; Women Engineering Fellow at Google; founder—Peer Programming Hub; prior internships (Swiggy, Koo, HoVAR); strong Linux/OS, DSA.

Interview Process They Faced (how many rounds, what type)

• Recruiter screening (availability, stack).
• Telephonic coding (Capacity & Performance Engg.) – 45 min on Zoom/CoderPad.
• Online Assessment (Production Engg.) – 20 Linux MCQs, 18 minutes.
• Telephonic coding (Production Engg.) – 45 min, 2 easy–medium coding problems.
• Final Systems Round – 45 min viva on OS/Linux administration concepts.

Questions Asked (statistical, ML, coding, case studies)

• Coding: string/stack manipulation, additional easy–medium problem solving on CoderPad.
• Linux/Systems: processes, OS fundamentals, admin tasks; practical Linux troubleshooting scenarios.
• General culture chat about Meta practices and team environment.

Outcome & Difficulty Level (cleared/not, feedback):

Not selected after systems round. Difficulty: moderate to high (time pressure in coding; depth expected in Linux/OS). Feedback implicit: strong problem understanding; improve coding speed and deep systems breadth.

Candidate Background (Experienced, skills, education):

Experienced backend/SWE; referral via Meta engineer; intensive prep in Java (collections, GC, concurrency), DSA (graphs/heaps/strings), and system design at scale.

Interview Process They Faced (how many rounds, what type)

• Intro call with recruiter (process, prep resources; month to prepare).
• Screening – fast-paced 45-min coding (arrays, trees—postorder, simple graph with in/out-degree).
• Finals (4 rounds):
• Coding – topo sort (hard), heap problem.
• Coding – modified binary search (optimal), DSU, quick string/sort task.
• System Design – Instagram feature: reqs, HLD, scalability, LLD (APIs, schema, client optimizations).
• Behavioral – Meta values: Focus on Impact, Move Fast, ownership, failure handling.

Questions Asked (statistical, ML, coding, case studies)

• Coding: graphs (topological ordering), heaps, binary search patterns, DSU; discuss complexities and alternatives.
• System Design: horizontally scalable social feature, non-functional requirements, cache/shard/partition, follow-ups on trade-offs and client/server optimizations.
• Behavioral: conflict resolution, mistakes, initiative; alignment with Meta principles.

Outcome & Difficulty Level (cleared/not, feedback):

Offer (E5, London). Recruiter shared three “strong hire” + one “hire” ratings. Difficulty: high (speed + depth). Key learnings: solve quickly, communicate trade-offs, keep DSA/system-design muscles sharp, and demonstrate ownership.