Backend Engineer • Behavioral • medium

Tell me about a time you strongly disagreed with a technical direction proposed by leadership. How did you handle it?

#Conflict Resolution #Communication #Ownership

Practice

Backend Engineer • Behavioral • medium

How do you prioritize technical debt versus shipping new features in a fast-paced environment?

#Prioritization #Technical Debt #Productivity

Practice

Backend Engineer • Behavioral • medium

Tell me about a time you caused a production outage. What was the root cause, and what did you learn?

#Accountability #Post-mortems #Learning

Practice

Backend Engineer • Behavioral • medium

Tell me about a time you had to dive into an undocumented, legacy codebase to fix a critical bug.

#Problem Solving #Adaptability #Debugging

Practice

Backend Engineer • Behavioral • medium

Tell me about a time you had to ship a critical feature under an extremely tight deadline. What trade-offs did you make?

#Delivery #Prioritization #Trade-offs

Practice

Backend Engineer • Coding • hard

Implement a basic LFU (Least Frequently Used) Cache.

#Hash Map #Linked List #Design

Practice

Backend Engineer • Coding • medium

Given an array of intervals representing active ad campaigns, merge all overlapping intervals to find the total continuous time ads are running.

#Array #Sorting #Intervals

Practice

Backend Engineer • Coding • medium

Given a stream of tweets, write an algorithm to find the top K trending hashtags over the last 1 hour.

#Heap #Hash Map #Sliding Window

Practice

Backend Engineer • Coding • medium

Given a list of user connections (follower/following), write a function to find the shortest connection path between two users (Degrees of Separation).

#Graph #BFS #Bidirectional BFS

Practice

Backend Engineer • Coding • hard

Write an algorithm to serialize and deserialize a tree of threaded replies (N-ary tree) to a string and back.

#Tree #DFS #Serialization

Practice

Backend Engineer • Coding • medium

Implement a Trie (Prefix Tree) to support autocomplete for user handles (@mentions).

#Trie #Design #String

Practice

Backend Engineer • Coding • medium

Given a list of tweets with timestamps, find the maximum number of tweets published within any 5-minute sliding window.

#Sliding Window #Two Pointers #Array

Practice

Backend Engineer • Coding • easy

Given two arrays representing the followers of User A and User B, write a function to return their mutual followers. Optimize for large arrays.

#Array #Hash Set #Two Pointers

Practice

Backend Engineer • Coding • medium

Write a function to detect if there is a cycle in a retweet chain (represented as a directed graph).

#Graph #DFS #Cycle Detection

Practice

Backend Engineer • Coding • easy

Implement a function to parse a tweet string and extract all unique hashtags and user mentions, returning them in the order they appeared.

#String Manipulation #Regex #Hash Set

Practice

Backend Engineer • Coding • medium

Write a function to implement an LRU (Least Recently Used) Cache. This is often used for caching user profiles in memory.

#Linked List #Hash Map #Design

Practice

Backend Engineer • Coding • easy

Given a string containing a tweet, check if it is a palindrome, ignoring spaces, punctuation, and casing.

#String #Two Pointers

Practice

Backend Engineer • System Design • hard

Design a real-time spam/bot detection system that evaluates tweets as they are published.

#Stream Processing #Machine Learning #Rules Engine

Practice

Backend Engineer • System Design • medium

Design Twitter's URL shortener (t.co). Focus on high availability and analytics tracking.

#Hashing #Database Sharding #Analytics

Practice

Backend Engineer • System Design • hard

Design the backend for Community Notes. How do you handle concurrent edits, version history, and fast retrieval?

#Concurrency Control #Database Design #Eventual Consistency

Practice

Backend Engineer • System Design • hard

Design the Direct Message (DM) system. How do you handle real-time delivery, read receipts, and offline users?

#WebSockets #Pub/Sub #Database Design

Practice

Backend Engineer • System Design • hard

Design the 'For You' timeline. How do you mix algorithmic recommendations with tweets from people the user follows?

#Recommendation Systems #Fan-out #Machine Learning Infrastructure

Practice

Backend Engineer • System Design • medium

Design the 'View Count' feature for tweets that was recently rolled out. How do you ensure high throughput and low latency without overwhelming the database?

#Caching #Event Sourcing #Data Aggregation

Practice

Backend Engineer • System Design • hard

Design a distributed rate limiter for the X API to handle different tiers of users (Free, Basic, Pro) with varying limits.

#Distributed Systems #Rate Limiting #Redis

Practice

Backend Engineer • System Design • medium

Design a Distributed ID Generator (like Snowflake) to generate unique IDs for tweets at a massive scale.

#Distributed Systems #ID Generation #Concurrency

Practice

Backend Engineer • System Design • medium

Design a system to securely store and serve media (images/videos) uploaded by users.

#Blob Storage #CDN #Media Processing

Practice

Backend Engineer • System Design • hard

Design a notification system that sends push notifications to millions of users simultaneously when a major news event occurs.

#Push Notifications #Queueing #Scalability

Practice

Backend Engineer • Technical • medium

How would you optimize a slow PostgreSQL query that is fetching a user's timeline and joining multiple tables (users, tweets, media)?

#SQL #Indexing #Query Optimization

Practice

Backend Engineer • Technical • hard

How would you handle a 'Thundering Herd' problem when a celebrity with 100 million followers tweets, and millions of users try to reply or like it simultaneously?

#Concurrency #Caching #Load Balancing

Practice

Backend Engineer • Technical • hard

You are tasked with migrating a legacy monolithic service to microservices to reduce compute costs. How do you approach this with zero downtime?

#Microservices #Migration #Strangler Fig Pattern

Practice

Backend Engineer • Technical • hard

How do you handle eventual consistency in a distributed database? Give an example of where strict consistency is required vs where eventual consistency is acceptable at X.

#CAP Theorem #Consistency Models #Databases

Practice

Backend Engineer • Technical • medium

Explain how Kafka partitions work. If a consumer group is lagging behind the producer in a timeline ingestion topic, how do you troubleshoot and resolve it?

#Kafka #Message Queues #Troubleshooting

Practice

Backend Engineer • Technical • medium

You notice a sudden spike in 503 Service Unavailable errors on the Timeline service. Walk me through your troubleshooting steps.

#Incident Response #Monitoring #Debugging

Practice

Backend Engineer • Technical • medium

Discuss the trade-offs between using gRPC vs REST for internal microservice communication at X.

#Networking #gRPC #API Design

Practice

Backend Engineer • Technical • hard

Explain how garbage collection works in the JVM. How would you tune it for a high-throughput backend service?

#JVM #Garbage Collection #Performance Tuning

Practice

Cloud Engineer • Behavioral • medium

How do you handle taking ownership of a legacy system where the original creators have left the company, there is minimal documentation, and it frequently fails?

#Legacy Systems #Ownership #Resilience

Practice

Cloud Engineer • Behavioral • medium

X has a strong focus on efficiency. Tell me about a time you had to drastically reduce cloud infrastructure costs. What was your approach and the outcome?

#Cost Optimization #Impact #Problem Solving

Practice

Cloud Engineer • Behavioral • medium

Describe a situation where you had to push back on a product or feature deployment because of infrastructure stability or security concerns. How did you handle the conflict?

#Communication #Conflict Resolution #Ownership

Practice

Cloud Engineer • Behavioral • medium

X moves extremely fast and sometimes things break. Tell me about a time you had to deploy a critical fix or make a major infrastructure decision with incomplete information.

#Adaptability #Decision Making #High Pressure

Practice

Cloud Engineer • Behavioral • hard

Tell me about the most complex production outage you've ever debugged. What was the root cause, how did you find it, and what did you do to ensure it never happened again?

#Incident Management #Debugging #Post-mortems

Practice

Cloud Engineer • Behavioral • easy

Describe a time you automated a tedious operational task. How did you identify the opportunity, what did you build, and how did you measure the impact?

#Automation #Toil Reduction #Initiative

Practice

Cloud Engineer • Behavioral • medium

Tell me about a time you strongly disagreed with a senior engineer or architect on a system design choice. How did you articulate your point, and what was the resolution?

#Communication #Technical Debate #Teamwork

Practice

Cloud Engineer • Coding • medium

Write a script (Python/Boto3 or Bash/AWS CLI) to automate the detection and rotation of AWS IAM access keys older than 90 days across multiple accounts.

#Python #AWS IAM #Security #Automation

Practice

Cloud Engineer • Coding • medium

Write a Python script to parse a 50GB Nginx access log file and output the top 10 IP addresses making requests. Optimize for memory usage.

#Python #Log Parsing #Memory Management #Data Structures

Practice

Cloud Engineer • Coding • medium

Given a list of microservices and their dependencies (e.g., Service A depends on Service B), write an algorithm to determine a valid deployment order.

#Graphs #Topological Sort #Algorithms

Practice

Cloud Engineer • Coding • medium

Implement a rate limiter class in Python or Go using the Token Bucket algorithm. It should support multiple users identified by user_id.

#Python #Go #Rate Limiting #Concurrency

Practice

Cloud Engineer • Coding • easy

Write a Bash one-liner or short script to find and gracefully terminate all zombie processes on a Linux machine.

#Bash #Linux #Process Management

Practice

Cloud Engineer • Coding • medium

Given a list of server connection logs represented as [start_time, end_time], write a function to find the maximum number of concurrent connections and the time it occurred.

#Algorithms #Sweep Line #Sorting

Practice

Cloud Engineer • Coding • medium

Write a Terraform module to provision a secure AWS VPC with 2 public subnets, 2 private subnets, a NAT gateway, and strict security groups allowing only internal traffic to the private subnets.

#Terraform #AWS #Networking

Practice

Cloud Engineer • Coding • medium

Implement a function that takes an IPv4 address string and a CIDR block string (e.g., '192.168.1.5', '192.168.1.0/24') and returns a boolean indicating if the IP belongs to the CIDR block.

#Networking #Bit Manipulation #Python #Go

Practice

Cloud Engineer • System Design • medium

Architect a disaster recovery plan for X's core authentication service. How do you ensure RTO (Recovery Time Objective) is under 1 minute and RPO (Recovery Point Objective) is zero?

#Disaster Recovery #High Availability #Databases #Failover

Practice

Cloud Engineer • System Design • hard

X has recently moved several workloads from AWS back to on-prem bare metal to save costs. Walk me through how you would architect the migration of a massive stateful service (like a distributed database) from AWS to an on-prem data center with zero downtime.

#Cloud Migration #Bare Metal #Database Replication #Networking

Practice

Cloud Engineer • System Design • hard

Design a global rate-limiting service for the X API that can handle millions of requests per second across multiple geographic regions.

#Rate Limiting #Distributed Systems #Redis #Concurrency

Practice

Cloud Engineer • System Design • hard

How would you design a highly available, multi-region Kubernetes cluster architecture to host X's core microservices?

#Kubernetes #High Availability #Multi-region #Service Mesh

Practice

Cloud Engineer • System Design • medium

Design a cost-optimized storage tiering system for X's media assets (images and videos) that balances retrieval speed with AWS S3/GCS storage costs.

#Storage #Cost Optimization #AWS S3 #CDN

Practice

Cloud Engineer • System Design • hard

Design an infrastructure deployment pipeline capable of deploying a critical security patch to 10,000+ nodes in under 5 minutes without causing a global outage.

#CI/CD #Automation #Deployment Strategies #Infrastructure as Code

Practice

Cloud Engineer • System Design • hard

Design the cloud infrastructure to support Twitter's timeline generation at scale, ensuring sub-200ms latency for 500 million daily active users.

#Scalability #Caching #Load Balancing #Microservices

Practice

Cloud Engineer • System Design • medium

Design a distributed caching layer for trending topics that can handle extreme traffic spikes (e.g., during the Super Bowl or a major global news event).

#Caching #Redis #Memcached #Thundering Herd

Practice

Cloud Engineer • Technical • hard

Describe the process of tuning a Linux kernel for high-throughput network traffic, specifically to handle millions of concurrent TCP connections.

#Linux #Networking #Kernel Tuning #TCP/IP

Practice

Cloud Engineer • Technical • medium

You notice a sudden 500ms latency spike in a critical microservice running on EKS. Walk me through your exact troubleshooting steps.

#Kubernetes #Observability #Networking #Debugging

Practice

Cloud Engineer • Technical • hard

Describe how you would architect a Prometheus and Grafana monitoring stack to ingest and query metrics from a fleet of 50,000 servers efficiently.

#Prometheus #Grafana #Monitoring #Scalability

Practice

Cloud Engineer • Technical • medium

What are the engineering and financial tradeoffs between using managed cloud databases (like AWS RDS or GCP Spanner) versus self-hosting databases on bare metal servers?

#Cost Optimization #Databases #Bare Metal #Cloud Computing

Practice

Cloud Engineer • Technical • medium

How do you secure internal microservice-to-microservice communication across different cloud VPCs or hybrid environments?

#Security #Networking #mTLS #Service Mesh

Practice

Cloud Engineer • Technical • hard

Explain BGP routing. How does X use BGP to announce IP prefixes for its global CDN network?

#BGP #CDN #Routing #Networking

Practice

Cloud Engineer • Technical • medium

What happens at the OS and container level when a Kubernetes pod runs out of memory (OOMKilled)? How do you debug and prevent it?

#Kubernetes #Linux #Memory Management #Docker

Practice

Cloud Engineer • Technical • hard

How does Kubernetes handle network partitions between nodes? Explain the split-brain scenario and how etcd consensus prevents it.

#Kubernetes #Distributed Systems #etcd #Consensus Algorithms

Practice

Cloud Engineer • Technical • hard

How would you implement zero-downtime database schema migrations for a live service handling 100k queries per second?

#Databases #Migrations #High Availability

Practice

Cloud Engineer • Technical • easy

Explain the difference between an Application Load Balancer (ALB) and a Network Load Balancer (NLB). When would you explicitly choose an NLB for a service at X?

#Load Balancing #AWS #OSI Model

Practice

Cloud Engineer • Technical • medium

How do you manage Terraform state for a massive, multi-team infrastructure without causing locking bottlenecks or state file corruption?

#Terraform #State Management #Collaboration #CI/CD

Practice

Cloud Engineer • Technical • medium

Walk me through the lifecycle of a DNS query when a user types x.com, focusing specifically on how global load balancing and Anycast routing are involved.

#DNS #Load Balancing #Anycast #BGP

Practice

Data Engineer • Behavioral • medium

Tell me about a time you had to ship a critical data pipeline under an extremely tight, almost impossible deadline.

#Time Management #Prioritization #Hardcore Work Ethic

Practice

Data Engineer • Behavioral • medium

Describe a situation where you identified a massive cost inefficiency in cloud infrastructure and the steps you took to fix it.

#Cost Optimization #Proactivity #Cloud Architecture

Practice

Data Engineer • Behavioral • medium

How do you handle working in an environment with high ambiguity, minimal documentation, and rapidly changing product requirements?

#Adaptability #Ambiguity #Communication

Practice

Data Engineer • Behavioral • medium

Tell me about a time you disagreed with a senior engineer about a technical architecture decision. How was it resolved?

#Conflict Resolution #Technical Communication #Ego

Practice

Data Engineer • Behavioral • medium

Walk me through a time when a data pipeline you owned failed in production, causing downstream impact. How did you debug and resolve it?

#Incident Management #Debugging #Accountability

Practice

Data Engineer • Coding • easy

Given a list of trending search terms, group the anagrams together.

#Strings #Hash Maps

Practice

Data Engineer • Coding • medium

Parse a log file of Twitter events to find the top 10 most active users in a given 1-hour window.

#Log Parsing #Hash Maps #Sorting #Priority Queue

Practice

Data Engineer • Coding • medium

Implement a sliding window algorithm to count the number of tweets containing a specific hashtag in the last 5 minutes.

#Sliding Window #Queues #Real-time Processing

Practice

Data Engineer • Coding • hard

Given K sorted streams of tweet IDs (in chronological order), merge them into a single sorted stream.

#Heaps #Pointers #Stream Processing

Practice

Data Engineer • Coding • hard

Find the shortest path between two users in the Twitter follower graph.

#Graphs #BFS #Bidirectional Search

Practice

Data Engineer • Coding • medium

Implement a rate limiter for the Twitter API using a token bucket algorithm.

#Concurrency #System Design #Object-Oriented Design

Practice

Data Engineer • Coding • medium

Design an algorithm to find the Top K frequent words in a continuous stream of tweets (Heavy Hitters problem).

#Count-Min Sketch #Heaps #Streaming Algorithms

Practice

Data Engineer • Coding • medium

Implement an LRU Cache to store recently accessed user profiles.

#Linked Lists #Hash Maps #Caching

Practice

Data Engineer • Coding • medium

Implement a Trie data structure to support Twitter search autocomplete.

#Trees #Trie #String Manipulation

Practice

Data Engineer • Coding • easy

Write a function to validate a JSON payload representing a Tweet object, ensuring all required fields are present and correctly typed.

#JSON #Type Checking #Error Handling

Practice

Data Engineer • System Design • hard

Design an ad-click attribution pipeline that handles late-arriving events and ensures exactly-once processing.

#Exactly-Once Semantics #Watermarks #Data Pipelines

Practice

Data Engineer • System Design • hard

How would you architect the migration of a massive on-premise Hadoop cluster to GCP BigQuery with zero downtime?

#Cloud Migration #BigQuery #Dual Writes #Data Validation

Practice

Data Engineer • System Design • hard

Design a real-time trending topics system capable of processing millions of tweets per second.

#Heavy Hitters #Stream Processing #Distributed Systems

Practice

Data Engineer • System Design • hard

Design the data pipeline for Twitter's View Count feature, ensuring real-time updates and high throughput.

#Stream Processing #Kafka #Redis #Event Sourcing

Practice

Data Engineer • System Design • medium

Design a relational data model for Twitter Spaces analytics, tracking hosts, listeners, and duration.

#Entity-Relationship #Normalization #Fact/Dimension Tables

Practice

Data Engineer • System Design • hard

How would you design the batch and streaming data pipelines to generate features for the 'For You' timeline recommendation engine?

#Feature Store #Lambda Architecture #Graph Processing

Practice

Data Engineer • System Design • hard

Design a data lake architecture for storing, partitioning, and querying 10PB of daily tweet logs efficiently.

#Data Lake #Partitioning #Parquet #Iceberg/Hudi

Practice

Data Engineer • System Design • hard

Architect a streaming system to detect spam and bot activity in real-time as tweets are published.

#Machine Learning Pipelines #Real-time Streaming #Feature Engineering

Practice

Data Engineer • Technical • medium

How does Kafka handle message ordering, and how would you ensure ordered processing of a single user's tweets across partitions?

#Kafka #Partitioning #Message Ordering

Practice

Data Engineer • Technical • hard

How would you handle exactly-once processing semantics in a Kafka to BigQuery streaming pipeline?

#Exactly-Once #Kafka #BigQuery #Idempotency

Practice

Data Engineer • Technical • hard

Explain how you would optimize a PySpark job that is suffering from severe data skew due to a viral tweet from Elon Musk.

#Spark #Data Skew #Salting #Broadcast Joins

Practice

Data Engineer • Technical • hard

Write a SQL query to calculate the conversion rate of ad impressions to clicks within a 1-hour window for each ad campaign.

#Time-based Joins #Aggregations #Performance Tuning

Practice

Data Engineer • Technical • medium

Write a SQL query to find users who have retweeted a specific tweet but do not follow the original author.

#Joins #Subqueries #Set Operations

Practice

Data Engineer • Technical • medium

Write a SQL query to find the top 3 trending hashtags per country on a given day using window functions.

#Window Functions #Ranking #CTEs

Practice

Data Engineer • Technical • medium

Write a SQL query to calculate the 7-day rolling average of tweets per user.

#Window Functions #Aggregations #Time Series

Practice

Data Engineer • Technical • easy

Explain the differences between Parquet and Avro file formats. When would you use each in our data ecosystem?

#File Formats #Parquet #Avro #Columnar vs Row-based

Practice

Data Engineer • Technical • medium

Compare Apache Flink and Spark Streaming. Which would you choose for calculating real-time engagement metrics at X, and why?

#Flink #Spark Streaming #Micro-batching vs Native Streaming

Practice

Data Engineer • Technical • medium

Given a table of user follows, write a SQL query to find all mutuals (users who follow each other).

#Self Joins #Filtering

Practice

Data Engineer • Technical • medium

Write a SQL query to find the median number of followers for users who joined X in 2023.

#Percentiles #Window Functions #Statistics

Practice

Data Engineer • Technical • easy

Write a SQL query to identify potential bots by finding users who tweeted more than 100 times in a single minute.

#GROUP BY #HAVING #Date Truncation

Practice

Data Scientist • Behavioral • medium

Tell me about a time you had to ship a data science model or analysis under an extremely tight deadline. What corners did you cut?

#Execution #Prioritization #Bias for Action

Practice

Data Scientist • Behavioral • medium

Why do you want to work at X, and how do you handle rapid, unexpected changes in company direction or product strategy?

#Adaptability #Resilience #Motivation

Practice

Data Scientist • Behavioral • easy

X operates with a very lean team and fast-paced environment. How do you prioritize your tasks when everything is labeled 'high priority'?

#Time Management #Impact #Adaptability

Practice

Data Scientist • Behavioral • easy

Explain the concept of a p-value to a non-technical Product Manager who wants to launch a feature based on an A/B test.

#Statistics #Stakeholder Management

Practice

Data Scientist • Behavioral • medium

Describe a situation where you disagreed with a product manager or engineering lead on a key metric. How did you resolve it?

#Conflict Resolution #Data-Driven #Leadership

Practice

Data Scientist • Behavioral • medium

Tell me about a time your data analysis proved a widely held assumption within your company wrong.

#Analytical Thinking #Influence #Communication

Practice

Data Scientist • Coding • hard

Write a SQL query to calculate the Day 1 and Day 7 retention rates for users who recently subscribed to X Premium.

#Cohort Analysis #Self Joins #Date Functions

Practice

Data Scientist • Coding • medium

Write a SQL query to calculate the 3-day rolling average of tweet impressions for each user over the past month.

#Window Functions #Aggregation #Time Series

Practice

Data Scientist • Coding • medium

Write a SQL query to find the top 3 creators by ad revenue payout in each geographic region for the last quarter.

#Ranking #Window Functions #Joins

Practice

Data Scientist • Coding • easy

Write a Python function to parse and extract all unique trending hashtags from a massive stream of tweet texts efficiently.

#Regex #Strings #Data Structures

Practice

Data Scientist • Coding • easy

Write a SQL query to find users who liked a specific tweet but do not follow the author of that tweet.

#Joins #Filtering #Set Operations

Practice

Data Scientist • Coding • medium

Given a dataset of user sessions, write a Python script to merge overlapping session time intervals.

#Sorting #Intervals #Arrays

Practice

Data Scientist • Coding • hard

Write a SQL query to calculate the median time between a user's account creation and their first tweet.

#Percentiles #Date Math #CTEs

Practice

Data Scientist • Coding • easy

Write a Python function to calculate the cosine similarity between two sparse user-feature vectors.

#Math #Arrays #Linear Algebra

Practice

Data Scientist • Coding • medium

Write a SQL query to find the percentage of tweets that receive at least one reply within 5 minutes of being posted.

#Date/Time Functions #Joins #Aggregation

Practice

Data Scientist • Coding • medium

Given an array of daily tweet counts for a specific hashtag, write a Python function to find the longest contiguous streak of days where the count strictly increased.

#Arrays #Dynamic Programming

Practice

Data Scientist • Coding • hard

Write a SQL query to identify 'bot rings'—groups of 5 or more users who have retweeted the exact same set of 10 tweets within a 1-hour window.

#Complex Joins #Grouping #Anomaly Detection

Practice

Data Scientist • Coding • medium

Implement a basic TF-IDF algorithm from scratch in Python for a small corpus of tweets.

#NLP #Hash Maps #Math

Practice

Data Scientist • Coding • hard

Write a SQL query to calculate the 'viral coefficient' of a tweet (average number of new retweets generated by each retweet).

#Graph Data #Recursive CTEs #Aggregation

Practice

Data Scientist • System Design • hard

Design a recommendation system for the 'For You' timeline. How do you balance chronological relevance with algorithmic personalization?

#Recommender Systems #Ranking #Two-Tower Models

Practice

Data Scientist • System Design • hard

Design a Graph ML system to power the 'Who to Follow' recommendations.

#Graph Neural Networks #Link Prediction #Scalability

Practice

Data Scientist • System Design • medium

How would you design a system to rank 'Trending Topics' in real-time?

#Ranking #Time Decay #NLP

Practice

Data Scientist • System Design • hard

How would you build a machine learning model to detect spam or bot accounts in real-time as they register or tweet?

#Anomaly Detection #Streaming #Classification

Practice

Data Scientist • Technical • medium

How would you deal with highly skewed data, such as user follower counts, when building a regression model?

#Data Transformation #Outliers #Modeling

Practice

Data Scientist • Technical • medium

How would you predict user churn for X Premium subscribers? What features would be most important?

#Classification #Survival Analysis #Feature Engineering

Practice

Data Scientist • Technical • hard

How do you evaluate the trade-off between increasing ad load in the timeline and potential degradation of user engagement?

#Trade-offs #Monetization #Experimentation

Practice

Data Scientist • Technical • hard

When would you choose a Multi-Armed Bandit approach over traditional A/B testing for optimizing ad placements on X?

#Multi-Armed Bandit #Reinforcement Learning #Optimization

Practice

Data Scientist • Technical • medium

We noticed a sudden 10% drop in Daily Active Users (DAU) on X. Walk me through how you would investigate the root cause.

#Debugging #Analytics #Product Sense

Practice

Data Scientist • Technical • hard

How would you optimize the creator ad revenue sharing model to ensure fairness while maximizing overall platform content creation?

#Optimization #Allocation #Economics

Practice

Data Scientist • Technical • medium

If a user has a 10% chance of seeing an ad on any given page load, what is the probability they see at least one ad in 5 page loads?

#Probability #Binomial Distribution

Practice

Data Scientist • Technical • hard

How would you design an A/B test to evaluate the impact of the 'Community Notes' feature on the spread of misinformation?

#A/B Testing #Metrics Definition #Misinformation

Practice

Data Scientist • Technical • medium

What metrics would you define to measure the success of the X Premium (formerly Twitter Blue) subscription service?

#Monetization #KPIs #Strategy

Practice

Data Scientist • Technical • medium

How would you build an NLP model to classify and hide highly toxic replies in a tweet thread?

#NLP #Classification #Trust & Safety

Practice

Data Scientist • Technical • hard

If we introduce long-form tweets, how do you account for the novelty effect in your A/B test analysis?

#Statistical Significance #Novelty Effect #User Behavior

Practice

Data Scientist • Technical • hard

How do you handle network effects and interference when running an A/B test on a highly connected social graph like X?

#Network Effects #Graph Clustering #A/B Testing

Practice

DevOps Engineer • Behavioral • hard

X moves extremely fast and sometimes breaks things. Tell me about a time you had to bypass standard procedures to ship something critical on a tight deadline.

#Agility #Risk Management #Decision Making

Practice

DevOps Engineer • Behavioral • medium

You are given 5 critical P0 issues at the exact same time during a major site-wide event. How do you prioritize and handle them?

#Prioritization #Incident Management #Communication

Practice

DevOps Engineer • Behavioral • medium

Tell me about a time you strongly disagreed with a technical decision made by leadership. How did you handle it, and what was the outcome?

#Conflict Resolution #Communication #Ownership

Practice

DevOps Engineer • Behavioral • medium

You receive an alert that API latency has spiked by 400% globally. Walk me through your incident response process from acknowledgment to resolution.

#SRE #Incident Response #On-call #Post-mortem

Practice

DevOps Engineer • Behavioral • medium

Tell me about a time you had to work extremely long hours to resolve a critical production outage. How did you handle the pressure and team dynamics?

#Resilience #Teamwork #High Pressure

Practice

DevOps Engineer • Coding • easy

Write a function to validate if a given string is a valid IPv4 or IPv6 address without using built-in IP validation libraries.

#String Manipulation #Networking #Validation

Practice

DevOps Engineer • Coding • easy

Write a script in Python or Bash to parse a massive Nginx access log file, extract all IP addresses, and return the top 10 IPs that encountered 5xx HTTP status codes.

#Bash #Python #Log Parsing #Regex

Practice

DevOps Engineer • Coding • hard

Implement a basic rate limiter in Python using a token bucket algorithm to protect our Tweet posting API from abuse.

#Python #Rate Limiting #Concurrency #Algorithms

Practice

DevOps Engineer • Coding • easy

Write a script to recursively find and delete files older than 30 days in a given directory, but keep files with a '.log' extension.

#Bash #Linux #File Management

Practice

DevOps Engineer • Coding • medium

Given a list of server logs with timestamps, write a program to find the peak traffic window of exactly 5 minutes.

#Sliding Window #Data Processing #Time Series

Practice

DevOps Engineer • Coding • medium

Write a script to query a Prometheus HTTP API endpoint and trigger a Slack webhook alert if CPU usage across a cluster exceeds 90% for 5 consecutive minutes.

#Python #Prometheus #API Integration #Alerting

Practice

DevOps Engineer • Coding • hard

Write a Terraform module to provision an AWS EKS cluster with managed node groups, VPC CNI, and IAM OIDC integration.

#Terraform #AWS #EKS #IAM

Practice

DevOps Engineer • Coding • medium

Write a Bash script to monitor disk usage. If any partition exceeds 85%, the script should find the top 5 largest directories in that partition and email the report.

#Bash #Linux #Monitoring

Practice

DevOps Engineer • System Design • hard

Design the caching layer for the X timeline using Redis to handle millions of concurrent reads and thousands of writes per second.

#Redis #Caching #High Throughput #Data Modeling

Practice

DevOps Engineer • System Design • hard

We are moving away from managed cloud services to self-hosted solutions to save money. How would you design a highly available, self-hosted Kafka cluster across 3 data centers?

#Kafka #Distributed Systems #Bare-metal #High Availability

Practice

DevOps Engineer • System Design • hard

Design a multi-region Kubernetes cluster architecture for X's timeline service to ensure 99.99% uptime, even if an entire region goes offline.

#Kubernetes #High Availability #Disaster Recovery #Global Load Balancing

Practice

DevOps Engineer • System Design • medium

How do you implement zero-downtime deployments for a stateless microservice receiving 100k requests per second?

#Deployment Strategies #Load Balancing #Kubernetes

Practice

DevOps Engineer • System Design • hard

We recently migrated a significant portion of our infrastructure from cloud back to bare-metal to optimize costs. Walk me through how you would architect the automated provisioning of 10,000 bare-metal servers across multiple data centers.

#Bare-metal #Automation #PXE #Ansible #Data Center

Practice

DevOps Engineer • System Design • hard

Explain how you would implement aggressive auto-scaling for stateless microservices to handle sudden viral events (like the Super Bowl) while minimizing idle compute costs.

#Auto-scaling #KEDA #Kubernetes #Cost Optimization

Practice

DevOps Engineer • System Design • medium

Design a monitoring and alerting stack for a newly launched live video streaming feature on X. What metrics are most critical?

#Prometheus #Grafana #Video Streaming #SLIs/SLOs

Practice

DevOps Engineer • Technical • medium

How does DNS resolution work? Walk me through the steps, and explain how you would configure DNS failover for a global service.

#DNS #Failover #Routing

Practice

DevOps Engineer • Technical • hard

X has a massive monorepo. How would you optimize our CI pipeline to reduce build and test times from 45 minutes to under 10 minutes?

#Bazel #Caching #Parallelization #Monorepo

Practice

DevOps Engineer • Technical • medium

We need to reduce our AWS/GCP cloud bill by $10M a month. Walk me through your strategy to identify and eliminate waste in a Kubernetes-heavy environment.

#FinOps #Kubernetes #Cloud Compute #Resource Requests/Limits

Practice

DevOps Engineer • Technical • medium

What exactly happens under the hood when you run `kubectl apply -f deployment.yaml`?

#Kubernetes Architecture #API Server #etcd #Kubelet #Controllers

Practice

DevOps Engineer • Technical • easy

How do you troubleshoot a Kubernetes pod that is stuck in CrashLoopBackOff, especially if the container logs are completely empty?

#Kubernetes #Debugging #Containers

Practice

DevOps Engineer • Technical • medium

A user complains that images on X are loading slowly in a specific geographic region (e.g., Southeast Asia). How do you troubleshoot this?

#CDN #DNS #Latency #BGP #Traceroute

Practice

DevOps Engineer • Technical • hard

Explain the TCP 3-way handshake. How would you tune TCP parameters on a Linux kernel to handle millions of high-throughput, low-latency connections?

#TCP/IP #Linux Kernel #Sysctl #Networking

Practice

DevOps Engineer • Technical • hard

How do you perform a schema migration on a massive PostgreSQL database table with billions of rows without locking the table or causing downtime?

#PostgreSQL #Database Migrations #Zero-downtime

Practice

DevOps Engineer • Technical • medium

How do you manage Terraform state files in a team of 50 engineers to prevent race conditions, state corruption, and security leaks?

#Terraform #State Management #Security

Practice

DevOps Engineer • Technical • easy

What is an inode? What happens when a Linux system runs out of inodes even if there is plenty of disk space left, and how do you fix it?

#Linux Filesystem #Inodes #Troubleshooting

Practice

DevOps Engineer • Technical • hard

How would you handle stateful sets in Kubernetes for a high-throughput, distributed database like Cassandra?

#StatefulSets #Cassandra #Persistent Volumes #Storage

Practice

DevOps Engineer • Technical • medium

Explain the difference between eventual consistency and strong consistency. Give an example of where you would use each within X's architecture.

#Distributed Systems #CAP Theorem #Databases

Practice

DevOps Engineer • Technical • medium

Describe your approach to managing secrets in a CI/CD pipeline. How do you prevent developers from accidentally hardcoding API keys?

#Secret Management #Vault #CI/CD #DevSecOps

Practice

DevOps Engineer • Technical • medium

A PostgreSQL database is experiencing 100% CPU utilization and extremely slow queries. How do you identify the root cause and resolve it?

#PostgreSQL #Performance Tuning #Troubleshooting

Practice

DevOps Engineer • Technical • medium

How do you calculate and enforce Error Budgets and SLOs for a critical service like the Tweet posting API?

#SLOs #Error Budgets #SRE Practices

Practice

Frontend Engineer • Behavioral • medium

How do you prioritize your tasks when faced with multiple high-severity production issues at the same time?

#Incident Management #Prioritization #Stress Management

Practice

Frontend Engineer • Behavioral • medium

Describe a time you had to learn a new technology or framework over a weekend to deliver a feature. How do you approach rapid learning?

#Adaptability #Continuous Learning #Drive

Practice

Frontend Engineer • Behavioral • medium

X moves at an incredibly fast pace. Give an example of how you adapt to rapidly changing requirements mid-sprint.

#Agility #Flexibility #Mindset

Practice

Frontend Engineer • Behavioral • hard

Tell me about a time you identified a major performance bottleneck in a frontend application and how you fixed it.

#Performance #Profiling #Optimization

Practice

Frontend Engineer • Behavioral • medium

Tell me about a time you had to ship a critical feature under an extremely tight deadline. How did you ensure quality while moving fast?

#Time Management #Prioritization #Execution

Practice

Frontend Engineer • Behavioral • medium

Describe a situation where you disagreed with a product manager or designer about a technical constraint. How did you resolve it?

#Communication #Conflict Resolution #Collaboration

Practice

Frontend Engineer • Behavioral • hard

Tell me about a time you took ownership of a failing project or a severely buggy codebase and turned it around.

#Ownership #Problem Solving #Resilience

Practice

Frontend Engineer • Coding • medium

Write a function that parses a raw tweet string, finds all @handles and #hashtags, and returns an array of objects representing the text segments and their types for rendering.

#Parsing #Regex #Data Transformation

Practice

Frontend Engineer • Coding • medium

Create an image carousel component. Implement lazy loading so that images are only fetched when they are one slide away from being visible.

#Component Design #Lazy Loading #State Management

Practice

Frontend Engineer • Coding • medium

Implement a rate limiter function in JavaScript. It should take a function, a limit (number of calls), and a time window, and drop any calls that exceed the limit within the window.

#Algorithms #Time Complexity #Closures

Practice

Frontend Engineer • Coding • easy

Write a polyfill for `Array.prototype.reduce`.

#Prototypes #Array Methods #Functional Programming

Practice

Frontend Engineer • Coding • easy

Implement a custom `useDebounce` React hook that takes a value and a delay, and returns the debounced value.

#Hooks #Timers #Closures

Practice

Frontend Engineer • Coding • medium

Write a React component that implements Infinite Scrolling from scratch without using third-party libraries.

#Intersection Observer #DOM APIs #Pagination

Practice

Frontend Engineer • Coding • medium

Build a Tweet character counter function that takes a string and returns the count, but URLs should always count as exactly 23 characters regardless of their actual length.

#String Manipulation #Regex #Edge Cases

Practice

Frontend Engineer • Coding • medium

Implement a custom `EventEmitter` class in JavaScript with `on`, `off`, `emit`, and `once` methods.

#Design Patterns #Data Structures #Closures

Practice

Frontend Engineer • Coding • hard

Write a function to deeply flatten a nested DOM tree into a single flat array of nodes, maintaining depth-first order.

#Recursion #Tree Traversal #DOM APIs

Practice

Frontend Engineer • Coding • medium

Implement `Promise.all()` from scratch. It should take an array of promises and return a single promise that resolves with an array of results, or rejects immediately if any promise rejects.

#Asynchronous Programming #Promises #Error Handling

Practice

Frontend Engineer • Coding • easy

Create a custom `usePrevious` hook in React that returns the value of a state variable from the previous render cycle.

#Hooks #Refs #Component Lifecycle

Practice

Frontend Engineer • Coding • medium

Write a function to deeply merge two JavaScript objects. If a key contains an array in both objects, concatenate them. If it's an object, merge recursively.

#Recursion #Object Manipulation #Type Checking

Practice

Frontend Engineer • Coding • medium

Implement a generic `memoize` function. Add a feature where the memoized values expire after a given Time To Live (TTL) in milliseconds.

#Caching #Closures #Time Complexity

Practice

Frontend Engineer • Coding • hard

Build a simple virtualized list component in React. It should accept an array of 100,000 items, a row height, and a viewport height, and only render the visible DOM nodes.

#Performance #DOM Manipulation #Math

Practice

Frontend Engineer • System Design • hard

Design the Twitter / X Home Timeline. Focus on the frontend architecture, data fetching strategy, and rendering performance for a feed with thousands of tweets containing rich media.

#Virtualization #Pagination #State Management #Performance

Practice

Frontend Engineer • System Design • hard

Design the Tweet Composer. How would you handle the 280-character limit, rich text highlighting (mentions, hashtags), media uploads, and optimistic UI updates upon submission?

#ContentEditable #Optimistic UI #File Uploads #Regex

Practice

Frontend Engineer • System Design • medium

Design the Twitter Search Autocomplete dropdown. How do you ensure high performance and prevent race conditions when a user types rapidly?

#Debouncing #Caching #Race Conditions #Accessibility

Practice

Frontend Engineer • System Design • medium

Design the Notifications dropdown. How would you architect the real-time updates for likes, retweets, and mentions?

#WebSockets #Server-Sent Events #State Synchronization

Practice

Frontend Engineer • System Design • hard

Design the Direct Messaging (DM) interface. Focus on offline support, real-time message delivery, and read receipts.

#WebSockets #IndexedDB #Offline First #Optimistic UI

Practice

Frontend Engineer • System Design • medium

Design a Poll widget for a Tweet. How do you handle state synchronization across multiple clients and calculate/animate percentages in real-time?

#Component Design #Animation #Data Synchronization

Practice

Frontend Engineer • System Design • medium

Design the 'Trending Topics' sidebar. Discuss your data fetching strategy and how you would keep the list fresh without overwhelming the server.

#Polling #Caching #Component Lifecycle

Practice

Frontend Engineer • Technical • medium

What are the security implications of rendering user-generated content (like Tweets) and how do you prevent XSS attacks in a modern frontend framework?

#XSS #Sanitization #Content Security Policy

Practice

Frontend Engineer • Technical • medium

Explain how you would implement accessibility (a11y) features for the Tweet composer, ensuring it is fully usable by screen readers and keyboard-only users.

#A11y #ARIA #Keyboard Navigation

Practice

Frontend Engineer • Technical • medium

Describe how you would implement optimistic UI updates when a user clicks the 'Like' button on a tweet, including error handling and rollback.

#Optimistic UI #Error Handling #Network Requests

Practice

Frontend Engineer • Technical • medium

How do you handle state management for a global entity like the 'current user' versus highly localized state like a 'tweet draft' in a large-scale React app?

#State Management #Redux #Context API

Practice

Frontend Engineer • Technical • medium

How would you architect a frontend application to support multiple themes (e.g., Light, Dim, Lights Out) dynamically without requiring a page reload?

#CSS Variables #Theming #Context

Practice

Frontend Engineer • Technical • medium

Explain how React 18's concurrent features (like useTransition and Suspense) can be used to improve the responsiveness of the Twitter feed.

#React 18 #Concurrency #UX

Practice

Frontend Engineer • Technical • hard

How would you optimize the Core Web Vitals (LCP, CLS, INP) specifically for a user's Twitter profile page?

#Core Web Vitals #Optimization #Rendering Strategies

Practice

Full Stack Engineer • Behavioral • hard

What is the most complex system you have built from scratch, and what were the key technical trade-offs you had to make?

#Architecture #Trade-offs #End-to-End Ownership

Practice

Full Stack Engineer • Behavioral • hard

Describe a time you had to take over a legacy codebase with no documentation and the original authors had left. How did you approach it?

#Adaptability #Problem Solving #Legacy Code

Practice

Full Stack Engineer • Behavioral • medium

X operates with a hardcore engineering culture. Tell me about a time you had to ship a critical feature under an extremely tight, almost unreasonable deadline.

#Time Management #Prioritization #Resilience

Practice

Full Stack Engineer • Behavioral • medium

Tell me about a time you identified a major performance bottleneck in your system and took the initiative to fix it without being asked.

#Initiative #Performance #Ownership

Practice

Full Stack Engineer • Behavioral • medium

How do you handle disagreements with a senior engineer or product manager regarding the technical direction of a project?

#Conflict Resolution #Communication #Teamwork

Practice

Full Stack Engineer • Behavioral • medium

X moves very fast, often requiring decisions without full context. Tell me about a time you had to make a technical decision with incomplete information.

#Decision Making #Ambiguity #Risk Management

Practice

Full Stack Engineer • Behavioral • hard

Describe a situation where you had to debug a severe production outage. What was your process from discovery to post-mortem?

#Incident Management #Debugging #Communication

Practice

Full Stack Engineer • Coding • medium

Given a stream of tweets, write an algorithm to find the Top K frequent words (Trending Topics) in the last hour.

#Heaps #Hash Maps #Stream Processing

Practice

Full Stack Engineer • Coding • medium

Implement a custom debounce function in JavaScript that supports both leading and trailing edge execution. This is used for the search bar on X.

#JavaScript #Closures #Asynchronous Programming

Practice

Full Stack Engineer • Coding • easy

Write a function to parse a raw tweet string and extract all valid hashtags, ignoring punctuation immediately following the tag.

#String Parsing #Regex

Practice

Full Stack Engineer • Coding • medium

Design and implement an LRU (Least Recently Used) Cache to store recently fetched tweets on the client side.

#Linked Lists #Hash Maps #Caching

Practice

Full Stack Engineer • Coding • easy

Implement a deep clone function for a complex state object in React, ensuring that nested objects, arrays, and Date objects are copied by value, not reference.

#JavaScript #Recursion #Object Manipulation

Practice

Full Stack Engineer • Coding • hard

Find the shortest path between two users on X (Degrees of Separation). You are given an API call `getFollowers(userId)`.

#Graph Traversal #BFS #Bidirectional BFS

Practice

Full Stack Engineer • Coding • medium

Implement an autocomplete system for @mentions. As the user types, return the top 5 matching usernames based on a combination of prefix match and the user's interaction frequency.

#Trie #Hash Maps #Sorting

Practice

Full Stack Engineer • Coding • hard

Given K sorted arrays representing timelines of users you follow, merge them into a single sorted timeline of the most recent tweets.

#Heaps #Merge Sort #Arrays

Practice

Full Stack Engineer • Coding • medium

Implement a rate limiter for the X API that allows a user to make N requests per time window T. How would you handle this in a distributed environment?

#Sliding Window #Concurrency #Distributed Systems

Practice

Full Stack Engineer • Coding • medium

Write a function to calculate the length of a Tweet. Note that URLs always count as 23 characters, and emojis/special Unicode characters should be counted correctly as single characters despite surrogate pairs.

#String Manipulation #Regex #Unicode

Practice

Full Stack Engineer • System Design • hard

Design a real-time Direct Messaging (DM) system for X, supporting 1-on-1 and group chats with read receipts.

#WebSockets #NoSQL #Real-time #Pub/Sub

Practice

Full Stack Engineer • System Design • hard

Design a system to detect and filter out spam or bot replies in real-time before they are rendered to users.

#Machine Learning #Stream Processing #Rules Engine

Practice

Full Stack Engineer • System Design • hard

Design an ad-serving system for the X feed. How do you select the most relevant ad in under 50ms?

#Machine Learning Inference #Low Latency #Auction Algorithms

Practice

Full Stack Engineer • System Design • hard

Design the architecture for X Spaces (live audio rooms). How do you handle thousands of listeners and multiple speakers with low latency?

#WebRTC #Audio Streaming #SFU #Scalability

Practice

Full Stack Engineer • System Design • medium

Design a search system to find specific tweets by a user within a specific date range.

#Elasticsearch #Inverted Index #Data Partitioning

Practice

Full Stack Engineer • System Design • medium

Design X's push notification system. How do you ensure notifications are delivered reliably and not duplicated?

#Message Queues #APNs/FCM #Idempotency

Practice

Full Stack Engineer • System Design • hard

Design a video upload and processing pipeline for X. Users can upload up to 2 hours of 1080p video.

#Blob Storage #Asynchronous Processing #CDN #Video Encoding

Practice

Full Stack Engineer • System Design • hard

Design the Like and Retweet counter system. How do you handle the massive write load when a tweet goes viral?

#Eventual Consistency #Batching #Distributed Counters

Practice

Full Stack Engineer • System Design • hard

Design X's Trending Topics system. How do you calculate trends in real-time across different geographic regions?

#Stream Processing #Apache Kafka #Count-Min Sketch #Distributed Systems

Practice

Full Stack Engineer • System Design • hard

Design the X Home Timeline. How do you ensure sub-second latency for users with millions of followers?

#Fanout #Caching #Message Queues #Database Sharding

Practice

Full Stack Engineer • Technical • hard

What strategies would you use to optimize the initial load time (Time to Interactive / First Contentful Paint) of the X web application?

#Performance #SSR #Code Splitting #Web Vitals

Practice

Full Stack Engineer • Technical • medium

Explain how you would implement optimistic UI updates for the 'Like' button on a tweet. How do you handle network failures?

#UX #State Management #Error Handling

Practice

Full Stack Engineer • Technical • medium

X uses GraphQL heavily. How does GraphQL help in reducing over-fetching in the web app compared to REST, and what are the caching challenges?

#GraphQL #REST #Caching

Practice

Full Stack Engineer • Technical • hard

Explain how you would handle state management for a complex web app like X Pro (formerly TweetDeck) with multiple independent columns updating in real-time.

#State Management #Redux #React Context #Architecture

Practice

Full Stack Engineer • Technical • medium

Describe the differences between Server-Sent Events (SSE) and WebSockets. Which would you choose for streaming new tweets to the client and why?

#WebSockets #SSE #HTTP

Practice

Full Stack Engineer • Technical • medium

How do you prevent Cross-Site Scripting (XSS) attacks when rendering user-generated content, such as tweets with links and formatting, in the browser?

#XSS #Security #Sanitization

Practice

Full Stack Engineer • Technical • medium

How would you optimize the rendering performance of an infinite scrolling list of tweets in React to prevent DOM bloat and memory leaks?

#React #Virtualization #Performance

Practice

Full Stack Engineer • Technical • hard

Explain database sharding strategies for storing billions of user relationships (followers/following). What is the shard key?

#Database Sharding #Scalability #RDBMS

Practice

Machine Learning Engineer • Behavioral • medium

Describe a time you strongly disagreed with a senior engineer or manager on a technical decision. How was it resolved?

#Conflict Resolution #Communication #Leadership

Practice

Machine Learning Engineer • Behavioral • medium

Twitter/X is known for a 'hardcore' engineering culture. Can you share an experience where you had to take extreme ownership of a failing project?

#Ownership #Resilience #Leadership

Practice

Machine Learning Engineer • Behavioral • medium

Tell me about a production outage you caused or helped resolve. What was the root cause and how did you prevent it from happening again?

#Incident Management #Accountability #Post-mortems

Practice

Machine Learning Engineer • Behavioral • medium

Tell me about a time you had to ship a complex ML feature under an extremely tight deadline. What corners did you cut?

#Delivery #Prioritization #Trade-offs

Practice

Machine Learning Engineer • Behavioral • medium

How do you prioritize fixing technical debt versus shipping new product features?

#Prioritization #Product Sense #Engineering Excellence

Practice

Machine Learning Engineer • Coding • medium

Given k sorted arrays of tweet timestamps, merge them into a single sorted array.

#Heaps #Divide and Conquer #Sorting

Practice

Machine Learning Engineer • Coding • hard

Given a stream of tweets, write a function to keep track of the median engagement score in a sliding window.

#Heaps #Sliding Window #Stream Processing

Practice

Machine Learning Engineer • Coding • medium

Implement a rate limiter for the Twitter API.

#System Design #Concurrency #Data Structures

Practice

Machine Learning Engineer • Coding • medium

Find the shortest path between two users in a massive social graph (degrees of separation).

#Graph Theory #BFS #Bidirectional Search

Practice

Machine Learning Engineer • Coding • medium

Design a data structure that supports insert, delete, and getRandom in O(1) time.

#Hash Tables #Arrays #Design

Practice

Machine Learning Engineer • Coding • medium

Given a stream of tweets, write an algorithm to find the Top K trending hashtags in the last 1 hour.

#Heaps #Sliding Window #Stream Processing

Practice

Machine Learning Engineer • Coding • easy

Design an algorithm to detect cycles in a directed graph (e.g., detecting retweet loops or bot rings).

#Graph Theory #DFS

Practice

Machine Learning Engineer • Coding • medium

Implement a Trie (Prefix Tree) to support insertion, search, and prefix search. This is used for hashtag autocomplete.

#Tries #Strings #Data Structures

Practice

Machine Learning Engineer • System Design • hard

Design the autocomplete and relevance ranking system for Twitter Search.

#Search #Information Retrieval #Tries #Ranking

Practice

Machine Learning Engineer • System Design • hard

Design an Ad Click-Through Rate (CTR) prediction model. How do you handle delayed feedback when users click an ad hours after viewing it?

#Ads Ranking #CTR Prediction #Data Pipelines

Practice

Machine Learning Engineer • System Design • medium

Design a system to recommend 'Who to Follow' to a user.

#Recommendation Systems #Graph ML #Batch Processing

Practice

Machine Learning Engineer • System Design • hard

Design a system to serve Grok (LLM) responses to users in real-time while maintaining the context of their recent tweets and timeline.

#LLMs #System Architecture #RAG #Real-time ML

Practice

Machine Learning Engineer • System Design • medium

Design a multimodal model architecture to classify NSFW content in tweets containing both text and images.

#Multimodal ML #Computer Vision #NLP #Classification

Practice

Machine Learning Engineer • System Design • hard

Design the 'For You' timeline ranking system. How do you balance engagement metrics like retweets and likes with negative signals like 'show less often'?

#Recommendation Systems #Ranking #Architecture #Real-time ML

Practice

Machine Learning Engineer • System Design • medium

Design a distributed counter system for Tweet views that can handle millions of concurrent updates.

#Distributed Systems #Scalability #Caching

Practice

Machine Learning Engineer • System Design • hard

Design a system to extract and rank trending topics in real-time across multiple languages.

#NLP #Stream Processing #System Architecture

Practice

Machine Learning Engineer • System Design • hard

How would you design a real-time bot detection system for Twitter?

#Anomaly Detection #Streaming Data #Classification #Graph ML

Practice

Machine Learning Engineer • Technical • hard

What techniques would you use to reduce the inference latency of a deep learning ranking model in production from 100ms to 20ms?

#Model Optimization #Inference #Efficiency

Practice

Machine Learning Engineer • Technical • medium

How do you handle severe class imbalance when training a spam detection model where spam is less than 0.1% of all tweets?

#Imbalanced Data #Classification #Loss Functions

Practice

Machine Learning Engineer • Technical • medium

Explain the difference between offline evaluation (e.g., NDCG, MAP) and online evaluation (A/B testing) for the home timeline. Why might they disagree?

#Evaluation Metrics #A/B Testing #Data Science

Practice

Machine Learning Engineer • Technical • hard

How would you use Reinforcement Learning to optimize long-term user engagement on the platform?

#Reinforcement Learning #Recommendation Systems #Optimization

Practice

Machine Learning Engineer • Technical • medium

How do you address position bias in the Twitter feed ranking model?

#Bias Mitigation #Ranking #Data Science

Practice

Machine Learning Engineer • Technical • hard

Explain how you would implement distributed training for a multi-billion parameter language model (like Grok).

#LLMs #Distributed Training #Deep Learning

Practice

Machine Learning Engineer • Technical • hard

What are the trade-offs between using FAISS (IVF-PQ) vs. HNSW for approximate nearest neighbor search in tweet retrieval?

#Vector Databases #ANN #Information Retrieval

Practice

Machine Learning Engineer • Technical • medium

How would you detect hate speech or toxic replies in real-time under strict latency constraints?

#NLP #Classification #Real-time ML #Efficiency

Practice

Machine Learning Engineer • Technical • hard

Explain the contrastive loss function used in training user-tweet embeddings. How do you select hard negatives?

#Loss Functions #Representation Learning #Embeddings

Practice

Machine Learning Engineer • Technical • hard

How do you evaluate a Generative LLM used for summarizing long Twitter threads or generating Grok responses?

#LLMs #Evaluation Metrics #NLP

Practice

Machine Learning Engineer • Technical • medium

Contrast Two-Tower models with Cross-Attention models. Why do we use Two-Tower for candidate generation and Cross-Attention for final ranking?

#Deep Learning #Information Retrieval #Model Architecture

Practice

Machine Learning Engineer • Technical • medium

How do you handle the cold start problem for new users in the 'For You' feed?

#Cold Start #Recommendation Systems #Heuristics

Practice

Machine Learning Engineer • Technical • hard

Explain how you would train a Graph Neural Network (GNN) on the Twitter follower graph to generate user embeddings.

#Graph Neural Networks #Embeddings #Distributed Training

Practice

Product Manager • Behavioral • hard

How do you manage engineering teams that are experiencing high burnout or turnover?

#Empathy #Team Management #Culture

Practice

Product Manager • Behavioral • medium

Give an example of how you used data to change a stubborn stakeholder's mind.

#Influence #Data-Driven #Communication

Practice

Product Manager • Behavioral • hard

Describe a time you strongly disagreed with executive leadership on a product direction. How did you handle it?

#Conflict Resolution #Communication #Data-Driven Decisions

Practice

Product Manager • Behavioral • medium

How do you prioritize features when your engineering resources are cut in half unexpectedly?

#Prioritization #Resource Management #Agility

Practice

Product Manager • Behavioral • medium

Tell me about a time you had to make a critical product decision with incomplete or ambiguous data.

#Decision Making #Ambiguity #Risk Management

Practice

Product Manager • Behavioral • medium

Tell me about a product or feature you launched that completely failed. What was the root cause, and what did you learn?

#Learning #Accountability #Self-Awareness

Practice

Product Manager • Behavioral • medium

Tell me about a time you had to ship a critical feature under an extremely tight deadline with limited resources.

#Execution #Resilience #Prioritization

Practice

Product Manager • Behavioral • medium

Tell me about a time you had to pivot a product strategy due to a sudden market change or competitor launch.

#Adaptability #Strategy #Agility

Practice

Product Manager • Coding • medium

Write a SQL query to find the top 5% of users by tweet volume over the last 30 days who are NOT subscribed to X Premium.

#Data Analysis #SQL #Monetization

Practice

Product Manager • Coding • medium

Write a SQL query to find the average time it takes for a Community Note to be published after a tweet is created.

#SQL #Performance Metrics #Data Analysis

Practice

Product Manager • Coding • medium

Write a SQL query to identify potential bot accounts based on tweet frequency (>100 per day) and a follower-to-following ratio of less than 0.1.

#Trust & Safety #SQL #Data Analysis

Practice

Product Manager • Coding • hard

Write a SQL query to calculate the week-over-week retention rate of Community Notes contributors.

#Retention #SQL #Cohort Analysis

Practice

Product Manager • System Design • medium

Design a rate-limiting system to prevent malicious bots from scraping X's API while allowing legitimate enterprise partners uninterrupted access.

#API #Security #Infrastructure

Practice

Product Manager • System Design • hard

Design a system to detect and filter out spam in X Direct Messages in real-time.

#Spam Detection #Real-time #Machine Learning

Practice

Product Manager • System Design • hard

Design a system to support X's trending topics. How do you ensure it is real-time, localized, and resistant to manipulation?

#Algorithms #Real-time Systems #Data Processing

Practice

Product Manager • System Design • hard

Design the backend for Community Notes to ensure real-time updates and consensus calculation across millions of users.

#Real-time #Distributed Systems #Algorithms

Practice

Product Manager • System Design • hard

Design the high-level architecture for X's 'For You' timeline recommendation engine.

#Machine Learning #Recommendation Systems #Scalability

Practice

Product Manager • System Design • hard

Design a system to handle viral spikes (e.g., Super Bowl, Election Night) to ensure X stays online and responsive.

#Scalability #Load Balancing #Infrastructure

Practice

Product Manager • System Design • hard

Design the backend architecture for X Spaces (live audio) to support 1 million concurrent listeners in a single room.

#Audio #Scalability #Real-time Systems

Practice

Product Manager • Technical • hard

How would you handle the product tradeoff between maximizing free speech and ensuring advertiser brand safety?

#Trade-offs #Strategy #Trust & Safety

Practice

Product Manager • Technical • medium

We want to increase long-form video consumption on X. What features would you build, and how would you measure success?

#Video #Engagement #Feature Development

Practice

Product Manager • Technical • medium

Evaluate the success of the Community Notes feature. What are the top 3 metrics you would look at?

#Trust & Safety #Metrics #Crowdsourcing

Practice

Product Manager • Technical • hard

Design a job search and recruiting feature (X Hiring) tailored specifically for the X platform.

#Market Expansion #Product Sense #B2B

Practice

Product Manager • Technical • hard

You run an A/B test for a new reply ranking algorithm. Engagement (replies/likes) goes up by 10%, but ad click-through rate (CTR) drops by 5%. Do you launch?

#Trade-offs #A/B Testing #Monetization

Practice

Product Manager • Technical • easy

How would you improve the Bookmarks feature on X to make it a core part of the daily user experience?

#User Experience #Core Product #Retention

Practice

Product Manager • Technical • hard

What do you believe is the biggest existential threat to X right now, and how would you counter it through product strategy?

#Competitive Analysis #Vision #Business Strategy

Practice

Product Manager • Technical • medium

How would you integrate Grok AI into the core X timeline experience to increase daily active user (DAU) engagement?

#AI #Product Sense #User Engagement

Practice

Product Manager • Technical • medium

Design an audio-first feature for X that is NOT live broadcasting (Spaces).

#Innovation #Audio #Asynchronous Communication

Practice

Product Manager • Technical • easy

What are the top 3 metrics you would track to evaluate the health of the X ad platform?

#Advertising #Metrics #Monetization

Practice

Product Manager • Technical • medium

How would you optimize the onboarding flow for new users to immediately understand X's value proposition?

#Growth #User Acquisition #UX

Practice

Product Manager • Technical • hard

Should X build a dating feature? Walk me through your strategic thinking.

#Product Sense #Vision #Market Expansion

Practice

Product Manager • Technical • medium

How would you design an analytics dashboard for X Creators? What are the most important data points to show?

#Analytics #UX #Creator Economy

Practice

Product Manager • Technical • medium

Pitch a new monetization feature for Creators on X that does not rely on ad-revenue sharing.

#Creator Economy #Monetization #Innovation

Practice

Product Manager • Technical • hard

X is transitioning into an 'Everything App'. Design a peer-to-peer payments feature integrated directly into the X platform.

#Fintech #Product Sense #Vision

Practice

Product Manager • Technical • medium

X Premium (formerly Twitter Blue) subscriptions have dropped by 15% week-over-week. Walk me through how you would investigate the root cause.

#Root Cause Analysis #Metrics #Monetization

Practice

Software Engineer • Behavioral • medium

Describe a situation where you strongly disagreed with a technical decision made by leadership. How did you handle it?

#Communication #Conflict Resolution #Ownership

Practice

Software Engineer • Behavioral • medium

Tell me about a time you had to ship a critical feature under an extremely tight deadline. How did you ensure quality?

#Delivery #Prioritization #Resilience

Practice

Software Engineer • Behavioral • hard

Describe a time you caused a production outage. What was the root cause, how did you fix it, and what systemic changes did you implement?

#Incident Management #Accountability #Post-mortems

Practice

Software Engineer • Behavioral • hard

Tell me about a time you had to dive into an unfamiliar, legacy codebase, understand it quickly, and optimize its performance.

#Problem Solving #Optimization #Adaptability

Practice

Software Engineer • Behavioral • medium

At X, we move incredibly fast and priorities shift. How do you manage your time when everything is labeled as a P0 (highest priority)?

#Time Management #Agility #Communication

Practice

Software Engineer • Coding • medium

Implement a thread-safe counter class for tracking tweet likes in a highly concurrent environment.

#Concurrency #Multithreading

Practice

Software Engineer • Coding • medium

Implement a sliding window algorithm to detect if a user has posted more than N tweets in the last M seconds.

#Queue #Sliding Window

Practice

Software Engineer • Coding • medium

Design an Autocomplete system for X handles. Implement `insert(handle)` and `search(prefix)`.

#Trie #String #Design

Practice

Software Engineer • Coding • easy

Given a tree representing a tweet thread (replies to replies), write a function to find the maximum depth of the thread.

#Tree #DFS #BFS

Practice

Software Engineer • Coding • easy

Given a string representing a tweet, calculate its length where URLs count as exactly 23 characters, regardless of their actual length.

#String #Parsing #Regex

Practice

Software Engineer • Coding • easy

You are given a list of user connections. Find the most influential user, defined as the user with the highest in-degree in the graph.

#Graph #Hash Table

Practice

Software Engineer • Coding • medium

Given a string, write a parser or state machine to extract all valid @mentions and #hashtags without using built-in regex libraries.

#String #State Machine

Practice

Software Engineer • Coding • medium

Implement an LRU Cache to store the most recently accessed tweets.

#Hash Table #Doubly Linked List #Design

Practice

Software Engineer • Coding • medium

Find the shortest path (degrees of separation) between two users in a massive directed graph of followers.

#Graph #BFS #Bidirectional Search

Practice

Software Engineer • Coding • hard

Given a stream of tweets, find the top K most frequent hashtags in the last 5 minutes.

#Sliding Window #Heap #Queue

Practice

Software Engineer • Coding • medium

Implement a Token Bucket rate limiter class with an `allow_request(user_id)` method.

#Concurrency #Math #Design

Practice

Software Engineer • Coding • hard

Given an array of k sorted arrays representing timelines from different users, merge them into a single sorted timeline.

#Heap #Divide and Conquer #Linked List

Practice

Software Engineer • Coding • medium

Design a simplified version of Twitter. Implement functions to post a tweet, get the 10 most recent tweets in a news feed, follow a user, and unfollow a user.

#Hash Table #Heap #Object-Oriented Design

Practice

Software Engineer • Coding • hard

Write a function to parse a massive log file of API requests and return the 99th percentile latency.

#Math #Sorting #Bucket Sort

Practice

Software Engineer • Coding • easy

Given an array of tweet engagement scores (positive and negative integers), find the contiguous subarray with the maximum sum.

#Array #Dynamic Programming

Practice

Software Engineer • System Design • hard

Design the X 'For You' timeline, focusing on how to handle the fan-out problem for users with millions of followers.

#Scalability #Caching #Graph Processing #Stream Processing

Practice

Software Engineer • System Design • medium

Design a distributed rate limiter for the X API to prevent abuse while allowing burst traffic for premium (X Premium) users.

#Distributed Systems #API Design #Concurrency

Practice

Software Engineer • System Design • hard

Design a push notification system that alerts millions of users within seconds when a highly followed account posts a new tweet.

#Pub/Sub #WebSockets #Mobile Push

Practice

Software Engineer • System Design • hard

Design X Search. How would you index tweets to allow full-text search within seconds of posting?

#Search Engines #Inverted Index #Real-time Indexing

Practice

Software Engineer • System Design • medium

Design the Community Notes feature, including the data model for contributors, ratings, and the algorithm pipeline to display the most helpful note.

#Data Modeling #Graph Algorithms #Event Driven

Practice

Software Engineer • System Design • hard

Design an Ad Click Aggregation system that processes billions of events per day and provides real-time analytics to advertisers.

#Big Data #OLAP #Exactly-once Semantics

Practice

Software Engineer • System Design • hard

Design a system to detect and mitigate bot swarms liking a specific tweet in real-time.

#Machine Learning Infrastructure #Anomaly Detection #Stream Processing

Practice

Software Engineer • System Design • hard

Design the Trending Topics system. How do you process millions of tweets per minute to find top hashtags in real-time?

#Stream Processing #Heavy Hitters #Data Pipelines

Practice

Software Engineer • System Design • medium

Design Snowflake: A distributed unique ID generator for tweets that is highly available and sortable by time.

#Distributed Systems #Bit Manipulation #High Availability

Practice

Software Engineer • System Design • hard

Design a View Count system for tweets that updates in near real-time and handles viral spikes without overwhelming the database.

#Event Aggregation #Caching #Database Sharding

Practice

Software Engineer • Technical • hard

How would you profile and reduce the memory footprint of a high-throughput Go or Rust microservice handling tweet ingestion?

#Profiling #Memory Management #Go/Rust

Practice

Software Engineer • Technical • easy

Describe the differences between TCP and UDP. Why might X use UDP for its Spaces live audio feature?

#Protocols #Networking #Streaming

Practice

Software Engineer • Technical • medium

Explain how a reverse proxy works. How does it help in load balancing and securing X's API traffic?

#Load Balancing #Security #Architecture

Practice

Software Engineer • Technical • medium

Explain how you would optimize a slow-loading React component on the X web client that renders a timeline of rich media tweets.

#React #Performance #Virtualization

Practice

Software Engineer • Technical • medium

How does Redis handle memory eviction? Which eviction policy would you configure for a tweet cache and why?

#Redis #Caching #Memory Management

Practice

Twitter / X

The Interview Loop

Recruiter Screen (30 min)

Technical Loop (3-4 Rounds)

Interview Question Bank

Tell me about a time you strongly disagreed with a technical direction proposed by leadership. How did you handle it?

How do you prioritize technical debt versus shipping new features in a fast-paced environment?

Tell me about a time you caused a production outage. What was the root cause, and what did you learn?

Tell me about a time you had to dive into an undocumented, legacy codebase to fix a critical bug.

Tell me about a time you had to ship a critical feature under an extremely tight deadline. What trade-offs did you make?

Implement a basic LFU (Least Frequently Used) Cache.

Given an array of intervals representing active ad campaigns, merge all overlapping intervals to find the total continuous time ads are running.

Given a stream of tweets, write an algorithm to find the top K trending hashtags over the last 1 hour.

Given a list of user connections (follower/following), write a function to find the shortest connection path between two users (Degrees of Separation).

Write an algorithm to serialize and deserialize a tree of threaded replies (N-ary tree) to a string and back.

Implement a Trie (Prefix Tree) to support autocomplete for user handles (@mentions).

Given a list of tweets with timestamps, find the maximum number of tweets published within any 5-minute sliding window.

Given two arrays representing the followers of User A and User B, write a function to return their mutual followers. Optimize for large arrays.

Write a function to detect if there is a cycle in a retweet chain (represented as a directed graph).

Implement a function to parse a tweet string and extract all unique hashtags and user mentions, returning them in the order they appeared.

Write a function to implement an LRU (Least Recently Used) Cache. This is often used for caching user profiles in memory.

Given a string containing a tweet, check if it is a palindrome, ignoring spaces, punctuation, and casing.

Design a real-time spam/bot detection system that evaluates tweets as they are published.

Design Twitter's URL shortener (t.co). Focus on high availability and analytics tracking.

Design the backend for Community Notes. How do you handle concurrent edits, version history, and fast retrieval?

Design the Direct Message (DM) system. How do you handle real-time delivery, read receipts, and offline users?

Design the 'For You' timeline. How do you mix algorithmic recommendations with tweets from people the user follows?

Design the 'View Count' feature for tweets that was recently rolled out. How do you ensure high throughput and low latency without overwhelming the database?

Design a distributed rate limiter for the X API to handle different tiers of users (Free, Basic, Pro) with varying limits.

Design a Distributed ID Generator (like Snowflake) to generate unique IDs for tweets at a massive scale.

Design a system to securely store and serve media (images/videos) uploaded by users.

Design a notification system that sends push notifications to millions of users simultaneously when a major news event occurs.

How would you optimize a slow PostgreSQL query that is fetching a user's timeline and joining multiple tables (users, tweets, media)?

How would you handle a 'Thundering Herd' problem when a celebrity with 100 million followers tweets, and millions of users try to reply or like it simultaneously?

You are tasked with migrating a legacy monolithic service to microservices to reduce compute costs. How do you approach this with zero downtime?

How do you handle eventual consistency in a distributed database? Give an example of where strict consistency is required vs where eventual consistency is acceptable at X.

Explain how Kafka partitions work. If a consumer group is lagging behind the producer in a timeline ingestion topic, how do you troubleshoot and resolve it?

You notice a sudden spike in 503 Service Unavailable errors on the Timeline service. Walk me through your troubleshooting steps.

Discuss the trade-offs between using gRPC vs REST for internal microservice communication at X.

Explain how garbage collection works in the JVM. How would you tune it for a high-throughput backend service?

How do you handle taking ownership of a legacy system where the original creators have left the company, there is minimal documentation, and it frequently fails?

X has a strong focus on efficiency. Tell me about a time you had to drastically reduce cloud infrastructure costs. What was your approach and the outcome?

Describe a situation where you had to push back on a product or feature deployment because of infrastructure stability or security concerns. How did you handle the conflict?

X moves extremely fast and sometimes things break. Tell me about a time you had to deploy a critical fix or make a major infrastructure decision with incomplete information.

Tell me about the most complex production outage you've ever debugged. What was the root cause, how did you find it, and what did you do to ensure it never happened again?

Describe a time you automated a tedious operational task. How did you identify the opportunity, what did you build, and how did you measure the impact?

Tell me about a time you strongly disagreed with a senior engineer or architect on a system design choice. How did you articulate your point, and what was the resolution?

Write a script (Python/Boto3 or Bash/AWS CLI) to automate the detection and rotation of AWS IAM access keys older than 90 days across multiple accounts.

Write a Python script to parse a 50GB Nginx access log file and output the top 10 IP addresses making requests. Optimize for memory usage.

Given a list of microservices and their dependencies (e.g., Service A depends on Service B), write an algorithm to determine a valid deployment order.

Implement a rate limiter class in Python or Go using the Token Bucket algorithm. It should support multiple users identified by user_id.

Write a Bash one-liner or short script to find and gracefully terminate all zombie processes on a Linux machine.

Given a list of server connection logs represented as [start_time, end_time], write a function to find the maximum number of concurrent connections and the time it occurred.

Write a Terraform module to provision a secure AWS VPC with 2 public subnets, 2 private subnets, a NAT gateway, and strict security groups allowing only internal traffic to the private subnets.

Implement a function that takes an IPv4 address string and a CIDR block string (e.g., '192.168.1.5', '192.168.1.0/24') and returns a boolean indicating if the IP belongs to the CIDR block.

Architect a disaster recovery plan for X's core authentication service. How do you ensure RTO (Recovery Time Objective) is under 1 minute and RPO (Recovery Point Objective) is zero?

X has recently moved several workloads from AWS back to on-prem bare metal to save costs. Walk me through how you would architect the migration of a massive stateful service (like a distributed database) from AWS to an on-prem data center with zero downtime.

Design a global rate-limiting service for the X API that can handle millions of requests per second across multiple geographic regions.

How would you design a highly available, multi-region Kubernetes cluster architecture to host X's core microservices?

Design a cost-optimized storage tiering system for X's media assets (images and videos) that balances retrieval speed with AWS S3/GCS storage costs.

Design an infrastructure deployment pipeline capable of deploying a critical security patch to 10,000+ nodes in under 5 minutes without causing a global outage.

Design the cloud infrastructure to support Twitter's timeline generation at scale, ensuring sub-200ms latency for 500 million daily active users.

Design a distributed caching layer for trending topics that can handle extreme traffic spikes (e.g., during the Super Bowl or a major global news event).

Describe the process of tuning a Linux kernel for high-throughput network traffic, specifically to handle millions of concurrent TCP connections.

You notice a sudden 500ms latency spike in a critical microservice running on EKS. Walk me through your exact troubleshooting steps.

Describe how you would architect a Prometheus and Grafana monitoring stack to ingest and query metrics from a fleet of 50,000 servers efficiently.

What are the engineering and financial tradeoffs between using managed cloud databases (like AWS RDS or GCP Spanner) versus self-hosting databases on bare metal servers?

How do you secure internal microservice-to-microservice communication across different cloud VPCs or hybrid environments?

Explain BGP routing. How does X use BGP to announce IP prefixes for its global CDN network?

What happens at the OS and container level when a Kubernetes pod runs out of memory (OOMKilled)? How do you debug and prevent it?

How does Kubernetes handle network partitions between nodes? Explain the split-brain scenario and how etcd consensus prevents it.

How would you implement zero-downtime database schema migrations for a live service handling 100k queries per second?

Explain the difference between an Application Load Balancer (ALB) and a Network Load Balancer (NLB). When would you explicitly choose an NLB for a service at X?

How do you manage Terraform state for a massive, multi-team infrastructure without causing locking bottlenecks or state file corruption?

Walk me through the lifecycle of a DNS query when a user types x.com, focusing specifically on how global load balancing and Anycast routing are involved.

Tell me about a time you had to ship a critical data pipeline under an extremely tight, almost impossible deadline.

Describe a situation where you identified a massive cost inefficiency in cloud infrastructure and the steps you took to fix it.

How do you handle working in an environment with high ambiguity, minimal documentation, and rapidly changing product requirements?

Tell me about a time you disagreed with a senior engineer about a technical architecture decision. How was it resolved?

Walk me through a time when a data pipeline you owned failed in production, causing downstream impact. How did you debug and resolve it?