AI Engineer • Behavioral • hard

Tell me about an AI project where you had to balance innovation with reliability.

#Reliability #Innovation

Practice

AI Engineer • Behavioral • medium

How do you handle stakeholder uncertainty around AI capabilities and limitations?

#Stakeholders #Expectations

Practice

AI Engineer • Behavioral • medium

Tell me about a time you optimized an LLM application for cost or latency.

#Cost #Latency

Practice

AI Engineer • Behavioral • medium

Describe a time you had to choose between using an AI model and a simpler rule-based system.

#Tradeoffs #Pragmatism

Practice

AI Engineer • Behavioral • easy

How do you stay current with the fast-moving AI/ML research landscape?

#Research #Continuous Learning

Practice

AI Engineer • Behavioral • hard

Tell me about a time an AI system you built produced unexpected or harmful outputs.

#Responsibility #Ethics

Practice

AI Engineer • Behavioral • hard

Describe an AI product you built from scratch. What were the key technical decisions?

#Product Development

Practice

AI Engineer • Behavioral • hard

Describe a situation where you had to debug a hard-to-reproduce AI model failure.

#Problem Solving

Practice

AI Engineer • Coding • hard

Implement a semantic chunking strategy for long documents.

#Chunking #Embeddings

Practice

AI Engineer • Coding • medium

Write a Python class to manage conversation history for a multi-turn chatbot.

#Chatbot #Memory

Practice

AI Engineer • Coding • hard

Implement a simple RAG pipeline using Python, LangChain, and FAISS.

#RAG #Python

Practice

AI Engineer • Coding • medium

Write a retry mechanism with exponential backoff for LLM API calls.

#Reliability #APIs

Practice

AI Engineer • System Design • hard

How would you architect an AI platform that supports 1000 concurrent LLM requests?

#Scaling #LLM Serving

Practice

AI Engineer • System Design • hard

Design an AI-powered customer support chatbot for an e-commerce platform.

#Chatbot #LLM

Practice

AI Engineer • System Design • hard

Design a document question-answering system using RAG.

#RAG #Vector Search

Practice

AI Engineer • System Design • hard

Design an AI code review system that integrates with GitHub PRs.

#Code Review #LLM

Practice

AI Engineer • System Design • hard

How would you build a multi-modal AI system that processes both text and images?

#Multi-Modal #Vision

Practice

AI Engineer • System Design • hard

Design a real-time AI safety filter for user-generated content.

#Content Moderation #Real-Time

Practice

AI Engineer • System Design • hard

Design an AI agent system that can autonomously browse the web and complete tasks.

#Agents #Tool Use

Practice

AI Engineer • Technical • medium

How do you choose the right embedding model for a domain-specific search task?

#Embedding Models #Search

Practice

AI Engineer • Technical • hard

Explain positional encoding in transformers. What are the differences between absolute and rotary position embeddings?

#Positional Encoding #RoPE

Practice

AI Engineer • Technical • hard

What is hallucination in LLMs? How do you detect and mitigate it?

#Hallucination #Safety

Practice

AI Engineer • Technical • medium

Explain the difference between autoregressive and masked language modeling.

#Autoregressive #Masked LM

Practice

AI Engineer • Technical • hard

What is a mixture of experts (MoE) architecture? How does it scale?

#MoE #Scaling

Practice

AI Engineer • Technical • hard

What is Constitutional AI? How does Anthropic use it?

#Constitutional AI #Anthropic

Practice

AI Engineer • Technical • hard

How do you red-team an AI system?

#Red Teaming #Security

Practice

AI Engineer • Technical • medium

What are guardrails in LLM applications? How do they work?

#Guardrails #Output Filtering

Practice

AI Engineer • Technical • medium

How do you integrate OpenAI API or Gemini API into a production application?

#OpenAI #Gemini

Practice

AI Engineer • Technical • medium

What is LangChain? What are its key components (Chains, Agents, Tools)?

#LangChain #Agents

Practice

AI Engineer • Technical • medium

What is streaming response from an LLM API? How do you implement it in a web app?

#Streaming #API

Practice

AI Engineer • Technical • medium

Explain structured output generation from LLMs (JSON mode, Instructor library).

#Structured Output #JSON

Practice

AI Engineer • Technical • hard

Explain how vector similarity search works. What are HNSW and IVF indices?

#HNSW #Similarity Search

Practice

AI Engineer • Technical • medium

Compare vector databases: Pinecone, Weaviate, Qdrant, and pgvector.

#Vector DB #Embeddings

Practice

AI Engineer • Technical • medium

What is semantic search? How does it differ from keyword-based search?

#Semantic Search #NLP

Practice

AI Engineer • Technical • hard

Explain the difference between dense and sparse retrieval in RAG.

#Dense Retrieval #BM25

Practice

AI Engineer • Technical • hard

How do you evaluate retrieval quality in a RAG system?

#Evaluation #Retrieval

Practice

AI Engineer • Technical • hard

How do you evaluate the quality of an LLM-generated response?

#LLM Evaluation #RAGAS

Practice

AI Engineer • Technical • hard

What is AI alignment? What are the key safety concerns with large-scale AI deployment?

#Alignment #Safety

Practice

AI Engineer • Technical • hard

Explain the concept of AI bias. How do you detect and mitigate it in production?

#Bias #Fairness

Practice

AI Engineer • Technical • medium

How do you manage LLM API rate limits and costs in production?

#Rate Limiting #Cost

Practice

AI Engineer • Technical • hard

Explain function calling / tool use in LLMs. How do you implement it?

#Function Calling #Tool Use

Practice

AI Engineer • Technical • hard

Explain the difference between GPT, BERT, and T5 architectures.

#GPT #BERT #T5

Practice

AI Engineer • Technical • medium

What is prompt engineering? What are few-shot, zero-shot, and chain-of-thought prompting?

#Prompt Engineering #Few-Shot

Practice

AI Engineer • Technical • hard

Explain how RLHF (Reinforcement Learning from Human Feedback) improves LLMs.

#RLHF #Alignment

Practice

AI Engineer • Technical • hard

What is RAG (Retrieval-Augmented Generation)? When would you use it over fine-tuning?

#RAG #Fine-Tuning

Practice

AI Engineer • Technical • medium

Explain the difference between fine-tuning and in-context learning.

#Fine-Tuning #ICL

Practice

AI Engineer • Technical • medium

What is token context window? How do you handle documents longer than the context limit?

#Context Window #Chunking

Practice

Cloud Engineer • Behavioral • medium

Tell me about a time you significantly reduced cloud infrastructure costs.

#FinOps #Impact

Practice

Cloud Engineer • Behavioral • medium

Describe a situation where a critical production system went down, and there was no runbook. How did you handle it?

#Incident Management #Ambiguity #Ownership #SRE

Practice

Cloud Engineer • Behavioral • medium

Tell me about a time you had to work with a difficult stakeholder or team member who strongly disagreed with your technical approach. How did you resolve it?

#Conflict Resolution #Communication #Teamwork #Influence

Practice

Cloud Engineer • Behavioral • medium

Tell me about a time you had to push back on a customer's architectural choice because you knew it would lead to scalability issues down the line.

#Customer Empathy #Communication #Pushback #Consulting

Practice

Cloud Engineer • Behavioral • medium

Describe your experience with incident post-mortems. What do you include?

#Post-Mortem #Learning

Practice

Cloud Engineer • Behavioral • medium

How do you communicate a complex cloud architecture to non-technical stakeholders?

#Stakeholders

Practice

Cloud Engineer • Behavioral • medium

Tell me about a time you improved the reliability of a cloud-based data system.

#SRE #Impact

Practice

Cloud Engineer • Behavioral • medium

Describe a situation where you had to choose between two cloud architectures. How did you decide?

#Architecture #Tradeoffs

Practice

Cloud Engineer • Behavioral • hard

Tell me about a major cloud outage you experienced. How did you respond?

#Outage #On-Call

Practice

Cloud Engineer • Behavioral • hard

Describe a time you migrated a critical workload to the cloud with zero downtime.

#Cloud Migration

Practice

Cloud Engineer • Behavioral • easy

How do you stay updated with new cloud services and features?

#Continuous Learning

Practice

Cloud Engineer • Coding • medium

Write a function to validate if a given string is a valid IPv4 address, and then extend it to check if it belongs to a specific CIDR block.

#String Manipulation #Bitwise Operations #Networking

Practice

Cloud Engineer • Coding • medium

Given a list of log entries with timestamps and error codes, write a function to find the top 3 most frequent error codes within a sliding window of 5 minutes.

#Sliding Window #Hash Map #Queue #Data Structures

Practice

Cloud Engineer • Coding • medium

Write a Python script to find and delete all unattached persistent disks in a GCP project that are older than 30 days to save costs.

#Python #GCP API #Cost Optimization #Scripting

Practice

Cloud Engineer • System Design • hard

How would you set up a streaming data pipeline on GCP using Pub/Sub and Dataflow?

#GCP #Pub/Sub #Dataflow

Practice

Cloud Engineer • System Design • hard

Design a data lake on AWS using S3, Glue, and Athena.

#AWS #S3 #Athena

Practice

Cloud Engineer • System Design • hard

Design a real-time streaming data pipeline on GCP to ingest, process, and analyze millions of IoT sensor events per second.

#Pub/Sub #Dataflow #BigQuery #IoT

Practice

Cloud Engineer • System Design • hard

Design a highly available, globally distributed web application on GCP that handles sudden, massive spikes in traffic (e.g., a viral news site).

#Global Load Balancer #Cloud CDN #Cloud Run #Cloud Spanner

Practice

Cloud Engineer • System Design • hard

A customer wants to migrate a monolithic on-premise application backed by an Oracle database to GCP. Walk me through your migration strategy.

#Cloud Migration #Strangler Fig #Database Migration Service #Bare Metal Solution

Practice

Cloud Engineer • System Design • hard

How do you implement disaster recovery for a cloud data warehouse?

#DR #RTO #RPO

Practice

Cloud Engineer • System Design • hard

How would you architect a data platform that reduces spend by 40% without impacting performance?

#FinOps #Cloud

Practice

Cloud Engineer • Technical • medium

What is the shared responsibility model in cloud security?

#Cloud Security #IAM

Practice

Cloud Engineer • Technical • hard

What is a VPC (Virtual Private Cloud)? How do you design a secure VPC architecture?

#VPC #Security

Practice

Cloud Engineer • Technical • easy

Explain the difference between regions, availability zones, and edge locations.

#Regions #AZs

Practice

Cloud Engineer • Technical • medium

How does auto-scaling work? What are the different scaling strategies?

#Auto-Scaling #EC2

Practice

Cloud Engineer • Technical • medium

What is a cloud-native application? How does it differ from a lifted-and-shifted one?

#Cloud Native #Migration

Practice

Cloud Engineer • Technical • hard

Explain multi-cloud vs hybrid cloud architectures and their tradeoffs.

#Multi-Cloud #Hybrid

Practice

Cloud Engineer • Technical • hard

Explain Kubernetes architecture: control plane, nodes, pods, and services.

#K8s #Containers

Practice

Cloud Engineer • Technical • hard

What is a Kubernetes Operator and when would you build one?

#Operators #CRD

Practice

Cloud Engineer • Technical • hard

How does container networking work in Kubernetes?

#Networking #CNI

Practice

Cloud Engineer • Technical • hard

Explain how you would design a cross-project IAM strategy for a large enterprise using Shared VPCs and least privilege principles.

#IAM #Shared VPC #Security #Resource Hierarchy

Practice

Cloud Engineer • Technical • hard

What happens exactly when you type `ls -l` in a Linux terminal? Go as deep into the OS level as possible.

#Linux #Syscalls #File Systems #Process Management

Practice

Cloud Engineer • Technical • medium

A customer complains that their GKE pods cannot reach an external API. Walk me through your troubleshooting steps.

#GKE #Networking #VPC #Cloud NAT

Practice

Cloud Engineer • Technical • medium

How would you implement a zero-downtime deployment strategy for a microservice running on Cloud Run?

#Cloud Run #CI/CD #Traffic Splitting #SRE

Practice

Cloud Engineer • Technical • medium

Explain the difference between a Readiness probe and a Liveness probe in Kubernetes. What happens if you misconfigure them?

#Kubernetes #GKE #Reliability #Microservices

Practice

Cloud Engineer • Technical • easy

Compare and contrast Cloud Storage, Persistent Disk, and Filestore. Give specific use cases for when you would choose one over the others.

#Storage #GCS #Block Storage #File Storage

Practice

Cloud Engineer • Technical • medium

What is OpenTelemetry? How does it standardize observability?

#OpenTelemetry #Tracing

Practice

Cloud Engineer • Technical • medium

How would you set up CloudWatch dashboards for a data pipeline?

#CloudWatch #AWS

Practice

Cloud Engineer • Technical • medium

Explain the three pillars of observability: logs, metrics, and traces.

#Logs #Metrics #Traces

Practice

Cloud Engineer • Technical • easy

What is a runbook? How do you create effective runbooks for data infrastructure?

#Runbook #On-Call

Practice

Cloud Engineer • Technical • medium

How do you do capacity planning for a cloud data platform?

#Scaling #Planning

Practice

Cloud Engineer • Technical • hard

Explain chaos engineering. How would you implement it for a data pipeline?

#Chaos Engineering #Fault Injection

Practice

Cloud Engineer • Technical • medium

What are SLOs, SLAs, and SLIs? How do you define them for a data platform?

#SLO #Reliability

Practice

Cloud Engineer • Technical • hard

How would you implement network segmentation for a multi-tier application?

#Security #Subnets

Practice

Cloud Engineer • Technical • medium

What is AWS PrivateLink? When would you use it?

#PrivateLink #VPC

Practice

Cloud Engineer • Technical • medium

How do cloud IAM roles and policies work? Explain least-privilege principle.

#IAM #Permissions

Practice

Cloud Engineer • Technical • medium

Explain TLS/SSL termination in a cloud load balancer.

#TLS #Load Balancer

Practice

Cloud Engineer • Technical • hard

What is zero-trust networking? How do you implement it on cloud?

#Zero Trust #Networking

Practice

Cloud Engineer • Technical • medium

How does AWS Glue Data Catalog work with Athena?

#Glue #Athena

Practice

Cloud Engineer • Technical • medium

Explain AWS S3 storage classes and lifecycle policies.

#S3 #Cost

Practice

Cloud Engineer • Technical • hard

What is BigQuery Slots? How do you optimize BigQuery query costs?

#GCP #Cost

Practice

Cloud Engineer • Technical • medium

Explain the difference between AWS Lambda and EC2 for data processing.

#Lambda #Serverless

Practice

Cloud Engineer • Technical • hard

Compare AWS EMR, GCP Dataproc, and Azure HDInsight for Spark workloads.

#EMR #Dataproc #Spark

Practice

Cloud Engineer • Technical • hard

How do you handle Terraform state across multiple teams?

#State Management #Collaboration

Practice

Cloud Engineer • Technical • medium

Explain idempotency in infrastructure provisioning.

#Idempotency #Terraform

Practice

Cloud Engineer • Technical • medium

How do you manage secrets in cloud infrastructure? (HashiCorp Vault, AWS Secrets Manager)

#Secrets Management #Vault

Practice

Cloud Engineer • Technical • medium

What is the difference between Terraform and Pulumi?

#Terraform #Pulumi

Practice

Cloud Engineer • Technical • hard

Explain Terraform's state management. What happens if the state file is corrupted?

#IaC #State

Practice

Cloud Engineer • Technical • medium

How does a Kubernetes Ingress controller work?

#Ingress #Load Balancing

Practice

Cloud Engineer • Technical • medium

Explain the difference between Docker and containerd.

#Docker #containerd

Practice

Cloud Engineer • Technical • hard

How would you set up horizontal pod autoscaling based on custom metrics?

#HPA #Custom Metrics

Practice

Cloud Engineer • Technical • hard

What is a service mesh? Explain how Istio works.

#Istio #Service Mesh

Practice

Cloud Engineer • Technical • medium

Explain Kubernetes resource requests vs limits. What happens if a pod exceeds its memory limit?

#Resources #OOM

Practice

Cloud Engineer • Technical • hard

Compare AWS, GCP, and Azure for a data-intensive workload. What are the key differentiators?

#AWS #GCP #Azure

Practice

Cloud Engineer • Technical • easy

Explain IaaS, PaaS, and SaaS with examples.

#IaaS #PaaS #SaaS

Practice

Data Analyst • Behavioral • medium

Tell me about an analysis that changed a major business decision.

#Business Impact #Influence

Practice

Data Analyst • Behavioral • medium

How do you handle a situation where a stakeholder challenges your analysis?

#Stakeholders #Confidence

Practice

Data Analyst • Behavioral • medium

Describe a time you found an insight that was counterintuitive.

#Curiosity

Practice

Data Analyst • Behavioral • hard

Tell me about a time you had incomplete data but still needed to deliver analysis.

#Ambiguity

Practice

Data Analyst • Behavioral • easy

How do you ensure your analyses are reproducible?

#Reproducibility

Practice

Data Analyst • Behavioral • medium

Tell me about a time you discovered data quality issues mid-analysis. What did you do?

#Problem Solving

Practice

Data Analyst • Behavioral • medium

How do you prioritize analytical requests when multiple teams need you?

#Time Management

Practice

Data Analyst • Behavioral • medium

Describe a dashboard you built that was widely adopted. What made it successful?

#Visualization

Practice

Data Analyst • Coding • hard

Write a SQL query to find customers who made purchases in both January and February but not March.

#Set Operations

Practice

Data Analyst • Coding • easy

Explain how groupby and agg work in pandas with an example.

#Pandas #GroupBy

Practice

Data Analyst • Coding • hard

What is a funnel query? Write one for a 3-step user onboarding flow.

#Funnel Analysis

Practice

Data Analyst • Coding • medium

Explain window functions. Write a query using LAG() to compute day-over-day change.

#Window Functions

Practice

Data Analyst • Coding • hard

Write a SQL query to calculate the rolling 28-day average session duration per user.

#Rolling Average #Sessions

Practice

Data Analyst • Coding • hard

How would you detect anomalies in a daily revenue time series using SQL?

#Anomaly Detection #SQL

Practice

Data Analyst • Coding • medium

What is a pivot table in SQL? How would you implement it without native PIVOT support?

#Pivot #Data Transformation

Practice

Data Analyst • Coding • medium

How would you merge two large DataFrames efficiently in pandas?

#Pandas #Merging

Practice

Data Analyst • Coding • medium

Describe how to detect and handle outliers in a dataset using Python.

#Outliers #Data Cleaning

Practice

Data Analyst • Coding • easy

Write Python code to load a CSV, clean missing values, and compute summary statistics.

#Data Cleaning #Pandas

Practice

Data Analyst • Coding • medium

Write a SQL query to calculate month-over-month revenue growth.

#Revenue #Growth Analytics

Practice

Data Analyst • Coding • hard

How would you build a cohort analysis for user retention in SQL?

#Cohort Analysis #Retention

Practice

Data Analyst • Coding • medium

How would you use pandas to compute a 7-day rolling average of sessions?

#Pandas #Time Series

Practice

Data Analyst • Technical • medium

Describe your process for creating an executive-level analytics presentation.

#Executive Reporting

Practice

Data Analyst • Technical • easy

How do you choose between a bar chart, line chart, and scatter plot?

#Charts #Design

Practice

Data Analyst • Technical • easy

Explain the difference between a HAVING clause and a WHERE clause.

#SQL Basics

Practice

Data Analyst • Technical • medium

How do you handle timezone conversions in SQL analytics?

#Timezones #Analytics

Practice

Data Analyst • Technical • hard

Daily Active Users dropped 15% yesterday. Walk me through how you'd investigate.

#Root Cause Analysis #Metrics

Practice

Data Analyst • Technical • medium

What is customer lifetime value (LTV)? How would you calculate it?

#LTV #Retention

Practice

Data Analyst • Technical • easy

Explain the difference between DAU, WAU, and MAU. Which is most useful and when?

#Engagement #KPIs

Practice

Data Analyst • Technical • medium

How would you measure the success of a new feature launch?

#Feature Success #Metrics

Practice

Data Analyst • Technical • easy

What is ARPU (Average Revenue Per User)? How do you segment ARPU analysis?

#ARPU #Revenue

Practice

Data Analyst • Technical • hard

Explain the concept of attribution modeling. What are last-click vs multi-touch models?

#Marketing Analytics

Practice

Data Analyst • Technical • medium

How would you build a dashboard to monitor e-commerce funnel health?

#Visualization #Funnel

Practice

Data Analyst • Technical • hard

What metrics would you use to measure the health of a marketplace?

#Marketplace #Supply & Demand

Practice

Data Analyst • Technical • easy

What is net promoter score (NPS)? How do you analyse NPS trends?

#NPS #Customer Satisfaction

Practice

Data Analyst • Technical • hard

How would you measure the impact of a pricing change on revenue?

#Pricing #A/B Test

Practice

Data Analyst • Technical • hard

Explain how you'd set up an A/B test to validate a new checkout flow.

#A/B Testing #Statistics

Practice

Data Analyst • Technical • hard

What sample size do you need for an A/B test? How do you calculate it?

#Sample Size #Power

Practice

Data Analyst • Technical • hard

A/B test shows p=0.04, but the effect size is tiny. Would you ship?

#Practical Significance #Decision Making

Practice

Data Analyst • Technical • medium

What is a novelty effect in experimentation? How do you account for it?

#Novelty Effect #Bias

Practice

Data Analyst • Technical • hard

How do you handle multiple metrics in an A/B test (metric tradeoffs)?

#Multiple Metrics #Tradeoffs

Practice

Data Analyst • Technical • medium

What makes a good data visualization? Walk me through your design principles.

#Design #Communication

Practice

Data Analyst • Technical • medium

How would you explain statistical significance to a non-technical product manager?

#Storytelling #Statistics

Practice

Data Analyst • Technical • easy

What tools do you use for dashboarding? Compare Tableau, Looker, and Metabase.

#Tableau #Looker

Practice

Data Engineer • Behavioral • medium

Describe a situation where you disagreed with a senior engineer or product manager on a technical design choice. How did you resolve it?

#Conflict Resolution #Collaboration #Googleyness

Practice

Data Engineer • Behavioral • medium

Tell me about a time you simplified a complex data platform decision across multiple teams.

#Communication #Stakeholders

Practice

Data Engineer • Behavioral • medium

Describe a situation where a data pipeline you owned went down in production. How did you handle it?

#On-Call #Problem Solving

Practice

Data Engineer • Behavioral • medium

How do you handle disagreements with data analysts or scientists who want features that compromise pipeline reliability?

#Conflict Resolution

Practice

Data Engineer • Behavioral • medium

Tell me about a time you significantly improved the performance of a data system.

#Performance #Optimization

Practice

Data Engineer • Behavioral • hard

Describe how you've balanced technical debt vs. new feature development in a data platform.

#Prioritization

Practice

Data Engineer • Behavioral • medium

Tell me about a time you onboarded a new data source that had significant quality issues.

#Problem Solving

Practice

Data Engineer • Behavioral • easy

Describe your experience mentoring junior data engineers.

#Mentoring #Collaboration

Practice

Data Engineer • Behavioral • easy

How do you stay current with rapidly evolving data engineering tools and practices?

#Growth Mindset

Practice

Data Engineer • Behavioral • medium

Tell me about a time you had to design a data pipeline with highly ambiguous requirements. How did you figure out what to build?

#Ambiguity #Googleyness #Communication

Practice

Data Engineer • Behavioral • medium

Tell me about a time a data pipeline you owned failed in production. What was the business impact, and what steps did you take to fix it and prevent it from happening again?

#Incident Management #Ownership #Post-mortem

Practice

Data Engineer • Coding • medium

Given a list of user session time intervals on YouTube represented as [start_time, end_time], write a Python function to merge all overlapping sessions and return the consolidated active viewing periods.

#Arrays #Sorting #Intervals

Practice

Data Engineer • Coding • hard

Implement a rate limiter in Python for an API using a sliding window approach. The rate limiter should allow a maximum of N requests per minute per user.

#Queues #Sliding Window #Concurrency

Practice

Data Engineer • Coding • medium

Write a SQL query to calculate the rolling 7-day average of daily video views per category on YouTube, ensuring days with zero views are still accounted for in the average.

#Window Functions #Aggregation #Rolling Averages

Practice

Data Engineer • Coding • medium

Write a Python function to parse a massive (100GB+) log file of Google Search queries and return the top K most frequent IP addresses. You have limited RAM.

#Heaps #Hash Maps #Log Parsing #Big O

Practice

Data Engineer • Coding • medium

Write a SQL query to find the top 3 highest-grossing apps in each region, but only include regions that have at least 100 active apps.

#CTEs #Window Functions #Filtering

Practice

Data Engineer • Coding • medium

Write a SQL query to find the second highest salary per department.

#Window Functions #SQL

Practice

Data Engineer • Coding • medium

Write a SQL query to compute a 7-day rolling average of daily sales.

#Window Functions #Analytics

Practice

Data Engineer • Coding • hard

Write a SQL query to find the longest streak of consecutive days a user has logged into Google Workspace. The input table has user_id and login_date.

#Window Functions #Gaps and Islands #CTEs

Practice

Data Engineer • System Design • hard

Design the data warehouse architecture for Google Play Store analytics. Stakeholders need daily reports on app downloads, revenue, and crash rates by region and device type.

#Data Warehousing #BigQuery #Schema Design #ETL

Practice

Data Engineer • System Design • hard

Design a data pipeline for Google Search query logs at 100K QPS.

#Scale #Google

Practice

Data Engineer • System Design • hard

Design a data model for an e-commerce platform tracking orders, users, and products.

#ER Modeling #Dimensional Modeling

Practice

Data Engineer • System Design • hard

Design an ETL pipeline that ingests 10TB of raw clickstream data daily.

#ETL #Batch Processing

Practice

Data Engineer • System Design • hard

How would you design a data pipeline that needs exactly-once delivery guarantees?

#Exactly-Once #Kafka

Practice

Data Engineer • System Design • hard

How would you design a real-time anomaly detection pipeline for 100K events/sec?

#Real-Time #Anomaly Detection

Practice

Data Engineer • System Design • hard

Design a real-time streaming data pipeline to detect click fraud in Google Ads. How would you ingest, process, and store the data to flag fraudulent clicks within seconds?

#Streaming #Pub/Sub #Dataflow #Fraud Detection

Practice

Data Engineer • System Design • hard

How would you design a data warehouse for a ride-sharing company from scratch?

#Architecture #Design

Practice

Data Engineer • System Design • hard

Design a batch processing pipeline to update Google Maps ETA prediction models based on daily historical traffic data. The data volume is petabytes per day.

#Batch Processing #MapReduce #DAGs #Orchestration

Practice

Data Engineer • Technical • medium

Explain the difference between S3, HDFS, and GCS for data storage.

#S3 #HDFS #GCS

Practice

Data Engineer • Technical • medium

Explain the concept of a data lakehouse. What are its advantages over a traditional data warehouse?

#Data Lakehouse #Data Warehouse

Practice

Data Engineer • Technical • hard

How do you handle schema evolution in a data pipeline without breaking downstream consumers?

#Schema Evolution #Backward Compatibility

Practice

Data Engineer • Technical • medium

What is a medallion architecture (Bronze/Silver/Gold)?

#Medallion #Data Lake

Practice

Data Engineer • Technical • medium

How do you implement data quality checks in a production pipeline?

#Great Expectations #Data Validation

Practice

Data Engineer • Technical • medium

What is data lineage and why is it important? How do you implement it?

#Lineage #Metadata

Practice

Data Engineer • Technical • hard

How would you detect and handle data drift in a production system?

#Data Drift #Monitoring

Practice

Data Engineer • Technical • medium

What is PII (Personally Identifiable Information) and how do you handle it in a data pipeline?

#PII #Privacy #Compliance

Practice

Data Engineer • Technical • medium

Explain the concept of a data catalog. What tools have you used?

#Data Catalog #Metadata

Practice

Data Engineer • Technical • hard

Compare AWS Redshift, Google BigQuery, and Snowflake for a petabyte-scale warehouse.

#Redshift #BigQuery #Snowflake

Practice

Data Engineer • Technical • hard

How does BigQuery handle large joins efficiently? What is its columnar storage approach?

#BigQuery #Columnar Storage

Practice

Data Engineer • Technical • medium

How would you reduce costs in a cloud-based data platform?

#Cloud #Cost

Practice

Data Engineer • Technical • medium

What is infrastructure as code (IaC)? Have you used Terraform for data infrastructure?

#Terraform #IaC

Practice

Data Engineer • Technical • hard

What is Data Vault methodology? How does it differ from Kimball?

#Data Vault #Kimball

Practice

Data Engineer • Technical • hard

You have a PySpark job running on Dataproc that joins a massive user table with a smaller transaction table. The job is taking hours and failing with OOM errors due to data skew. How do you optimize it?

#Spark #Data Skew #Salting #Performance Tuning

Practice

Data Engineer • Technical • medium

Explain the difference between partitioning and clustering in BigQuery. When would you use one over the other, and when would you use both?

#BigQuery #Partitioning #Clustering #Optimization

Practice

Data Engineer • Technical • medium

In Apache Beam or Google Cloud Dataflow, how do you handle late-arriving data in a windowed streaming pipeline?

#Streaming #Watermarks #Late Data #Apache Beam

Practice

Data Engineer • Technical • medium

What is the star schema vs snowflake schema? When would you use each?

#Star Schema #Snowflake Schema

Practice

Data Engineer • Technical • medium

Explain compaction in Delta Lake / Iceberg. Why is it important?

#Compaction #Performance

Practice

Data Engineer • Technical • hard

What is Delta Lake? How does it provide ACID transactions on data lakes?

#Delta Lake #ACID #Time Travel

Practice

Data Engineer • Technical • medium

Explain how Parquet and ORC file formats work and when you'd use each.

#Parquet #ORC #Columnar

Practice

Data Engineer • Technical • medium

What is the CAP theorem? Give an example of a real-world system tradeoff.

#CAP #Consistency #Availability

Practice

Data Engineer • Technical • medium

How does Kafka handle message ordering guarantees?

#Ordering #Partitions

Practice

Data Engineer • Technical • medium

What is Apache Kafka? Explain topics, partitions, consumer groups, and offsets.

#Kafka #Streaming

Practice

Data Engineer • Technical • hard

Explain the difference between map-side and reduce-side joins in MapReduce/Spark.

#Joins #MapReduce

Practice

Data Engineer • Technical • hard

What is data skew in Spark? How do you diagnose and fix it?

#Data Skew #Performance

Practice

Data Engineer • Technical • hard

Explain how Apache Spark's execution model works. What is a DAG in Spark?

#Spark #DAG #Distributed Computing

Practice

Data Engineer • Technical • easy

Explain the difference between push-based and pull-based data ingestion.

#Push #Pull #CDC

Practice

Data Engineer • Technical • medium

What is Apache Airflow? How does it differ from Prefect or Dagster?

#Airflow #Prefect #Dagster

Practice

Data Engineer • Technical • medium

How do you monitor data pipeline health in production? What metrics do you track?

#Monitoring #Alerting

Practice

Data Engineer • Technical • medium

Describe how you'd implement circuit breakers in a data pipeline.

#Circuit Breakers #Fault Tolerance

Practice

Data Engineer • Technical • hard

What is backfilling? How do you handle a backfill of 2 years of historical data without impacting production?

#Backfill #Airflow

Practice

Data Engineer • Technical • hard

Explain the Lambda architecture. What are its tradeoffs vs Kappa architecture?

#Lambda #Kappa #Streaming

Practice

Data Engineer • Technical • medium

What is idempotency and why is it critical in data pipelines?

#Idempotency #Data Quality

Practice

Data Engineer • Technical • hard

How do you handle late-arriving data in a streaming pipeline?

#Kafka #Watermarks

Practice

Data Engineer • Technical • medium

Explain ACID properties. Which databases sacrifice ACID for performance and why?

#ACID #Distributed Systems

Practice

Data Engineer • Technical • medium

What are CTEs (Common Table Expressions) and how do they differ from subqueries?

#CTEs #SQL

Practice

Data Engineer • Technical • hard

Describe partitioning strategies in a data warehouse. When would you use range vs hash partitioning?

#Partitioning #Performance

Practice

Data Engineer • Technical • medium

What is a materialized view? How does it differ from a regular view?

#Materialized Views #Performance

Practice

Data Engineer • Technical • medium

Explain the difference between RANK(), DENSE_RANK(), and ROW_NUMBER().

#Window Functions #SQL

Practice

Data Engineer • Technical • hard

How would you optimize a SQL query that is running slowly on a 1 billion row table?

#Query Optimization #Indexing

Practice

Data Engineer • Technical • hard

What is a slowly changing dimension (SCD)? Describe SCD Type 1, 2, and 3 with examples.

#SCD #Dimensional Modeling

Practice

Data Engineer • Technical • medium

Explain the difference between OLAP and OLTP systems. When would you use each?

#OLAP #OLTP #Databases

Practice

Data Engineer • Technical • hard

How would you model and optimize a BigQuery dataset for petabyte-scale ad-click attribution?

#BigQuery #Attribution

Practice

Data Engineer • Technical • hard

Explain how Google Spanner achieves global consistency with TrueTime.

#Spanner #TrueTime

Practice

Data Engineer • Technical • hard

How would you use Dataflow (Apache Beam) for a streaming aggregation job?

#Dataflow #Beam

Practice

Data Engineer • Technical • hard

What is Bigtable? When would you choose it over BigQuery?

#Bigtable #GCP

Practice

Data Engineer • Technical • medium

How do you optimize BigQuery costs for ad-hoc analytical queries?

#Cost #Optimization

Practice

Data Scientist • Behavioral • medium

Describe a project where you had to iterate significantly on your initial approach.

#Iteration #Learning

Practice

Data Scientist • Behavioral • medium

How do you prioritize between multiple data science requests from different teams?

#Stakeholder Management

Practice

Data Scientist • Behavioral • medium

How do you approach ethical considerations in ML model building?

#Fairness #Bias

Practice

Data Scientist • Behavioral • hard

Tell me about a time your model failed in production. What did you learn?

#Production #MLOps

Practice

Data Scientist • Behavioral • hard

Describe a time you used data to challenge a widely held assumption in your organization.

#Influence #Analytics

Practice

Data Scientist • Behavioral • medium

Tell me about a data science project where the results surprised you. What did you do?

#Analytical Thinking

Practice

Data Scientist • Behavioral • medium

Describe how you communicated a complex model result to a non-technical stakeholder.

#Storytelling

Practice

Data Scientist • Behavioral • hard

Tell me about a time you had to push back on a business request for an analysis that would be misleading.

#Ethics #Communication

Practice

Data Scientist • Coding • hard

Write a SQL query to calculate 30-day user retention.

#Retention #Analytics

Practice

Data Scientist • Coding • hard

How would you write a funnel analysis query in SQL?

#Funnel #Analytics

Practice

Data Scientist • Coding • medium

Write a query to identify duplicate records and deduplicate them.

#Deduplication #Data Quality

Practice

Data Scientist • System Design • hard

Design a feature store. What are its key components?

#Feature Store #MLOps

Practice

Data Scientist • System Design • hard

How would you build and deploy a churn prediction model?

#Churn #MLOps

Practice

Data Scientist • System Design • hard

Design a real-time fraud detection system for a payments platform.

#Fraud Detection #Real-Time ML

Practice

Data Scientist • System Design • hard

How would you build a recommendation system? Compare collaborative vs content-based filtering.

#Collaborative Filtering #Content-Based

Practice

Data Scientist • Technical • medium

How do you choose a north star metric for a product?

#Metrics #Product Strategy

Practice

Data Scientist • Technical • medium

How do you handle class imbalance in a classification problem?

#Imbalanced Data #SMOTE

Practice

Data Scientist • Technical • medium

What is regularization? Explain L1 vs L2 regularization and their effects.

#Regularization #L1 #L2

Practice

Data Scientist • Technical • medium

How does a Random Forest work? What are its hyperparameters and how do you tune them?

#Random Forest #Hyperparameter Tuning

Practice

Data Scientist • Technical • hard

Explain gradient boosting. How does XGBoost differ from a standard gradient boosting machine?

#Gradient Boosting #XGBoost

Practice

Data Scientist • Technical • medium

How would you detect and handle multicollinearity in a regression model?

#Multicollinearity #Regression

Practice

Data Scientist • Technical • hard

Explain the curse of dimensionality and its implications for ML models.

#Dimensionality #Feature Engineering

Practice

Data Scientist • Technical • medium

What is a confidence interval? How does it differ from a prediction interval?

#Confidence Interval #Intervals

Practice

Data Scientist • Technical • hard

Explain Bayesian vs Frequentist statistics. When would you use each?

#Bayesian #Frequentist

Practice

Data Scientist • Technical • hard

How do you evaluate the quality of a search ranking change at Google's scale?

#Search Ranking #Evaluation

Practice

Data Scientist • Technical • hard

How do you design an A/B test for a new product feature?

#A/B Testing #Statistics

Practice

Data Scientist • Technical • easy

What is the difference between Type I and Type II errors?

#Hypothesis Testing #Errors

Practice

Data Scientist • Technical • medium

Explain the central limit theorem and its importance in data science.

#CLT #Sampling

Practice

Data Scientist • Technical • medium

What is a p-value? Why is a p-value of 0.05 not always sufficient?

#Hypothesis Testing #p-value

Practice

Data Scientist • Technical • hard

Explain how Google's NDCG metric works for search relevance.

#NDCG #Relevance

Practice

Data Scientist • Technical • medium

Explain the bias-variance tradeoff. How does it influence model selection?

#Bias-Variance #Model Selection

Practice

Data Scientist • Technical • hard

What statistical techniques would you use to analyse Search CTR experiments?

#CTR #Statistics

Practice

Data Scientist • Technical • hard

What is the multiple testing problem? How do you correct for it?

#Bonferroni #FDR

Practice

Data Scientist • Technical • medium

Explain the ROC curve and AUC metric. When would you prefer AUC over accuracy?

#ROC #AUC #Metrics

Practice

Data Scientist • Technical • hard

What is a network effect in experimentation? How do you handle SUTVA violation?

#SUTVA #Network Effects

Practice

Data Scientist • Technical • hard

How would you design an experiment to measure the impact of a new ranking algorithm?

#Experimentation #Metrics

Practice

Data Scientist • Technical • medium

How would you detect and mitigate overfitting in a neural network?

#Overfitting #Dropout #Regularization

Practice

Data Scientist • Technical • medium

Explain batch normalization and why it helps training.

#Batch Normalization #Training

Practice

Data Scientist • Technical • medium

What is embedding? How do word embeddings like Word2Vec and GloVe work?

#Embeddings #Word2Vec

Practice

Data Scientist • Technical • medium

How would you approach an NLP problem like sentiment analysis from scratch?

#Sentiment Analysis #Text Classification

Practice

Data Scientist • Technical • medium

What is transfer learning? How would you fine-tune a pre-trained model?

#Transfer Learning #Fine-Tuning

Practice

Data Scientist • Technical • hard

Explain the transformer architecture. What are attention mechanisms?

#Transformers #Attention #BERT

Practice

Data Scientist • Technical • hard

What is the vanishing gradient problem? How do LSTM and ResNet address it?

#LSTM #ResNet #Gradients

Practice

Data Scientist • Technical • medium

Explain how backpropagation works.

#Backpropagation #Neural Networks

Practice

Data Scientist • Technical • medium

What is principal component analysis (PCA)? What are its limitations?

#PCA #SVD

Practice

Data Scientist • Technical • medium

Explain the difference between bagging and boosting.

#Bagging #Boosting

Practice

Data Scientist • Technical • medium

How do you approach feature selection?

#Feature Selection #LASSO

Practice

Data Scientist • Technical • medium

What is cross-validation? Explain k-fold and stratified k-fold.

#Cross Validation #k-Fold

Practice

Data Scientist • Technical • hard

How do you monitor model performance in production? What is model drift?

#Model Drift #Monitoring

Practice

Data Scientist • Technical • easy

Explain the difference between INNER JOIN, LEFT JOIN, and CROSS JOIN.

#Joins #SQL

Practice

Data Scientist • Technical • easy

What is an experiment holdout group?

#Holdout #Control Group

Practice

Data Scientist • Technical • hard

How would you identify the root cause of a sudden 20% drop in DAU?

#Root Cause Analysis #Debugging

Practice

Data Scientist • Technical • easy

Explain the difference between a leading indicator and a lagging indicator.

#Metrics #KPIs

Practice

Machine Learning Engineer • Behavioral • medium

Describe a situation where you had to push back on a Product Manager who wanted to launch an ML feature that achieved high accuracy but failed to meet the required P99 latency SLA.

#Conflict Resolution #Prioritization #Cross-functional Collaboration

Practice

Machine Learning Engineer • Behavioral • medium

Tell me about a time you discovered a significant bias or data leakage in your ML model right before deployment. How did you handle it, and how did you communicate the delay to stakeholders?

#Googleyness #Communication #Model Debugging #Ethics

Practice

Machine Learning Engineer • Coding • medium

Given a 2D grid representing a cluster of TPU v5e pods where '1' is an active pod and '0' is inactive, write an algorithm to find the maximum area of connected active pods. Pods are connected horizontally or vertically.

#Graph Theory #Depth-First Search #Breadth-First Search #Matrix

Practice

Machine Learning Engineer • Coding • medium

Implement a custom sparse matrix-vector multiplication (SpMV) algorithm. Assume the sparse matrix is provided in Compressed Sparse Row (CSR) format.

#Linear Algebra #Data Structures #Performance Optimization

Practice

Machine Learning Engineer • Coding • hard

Write a function to sample a random node from a massive, distributed graph where you only have access to an API `get_neighbors(node_id)`. You do not know the total number of nodes.

#Randomized Algorithms #Graph Theory #Reservoir Sampling #Markov Chains

Practice

Machine Learning Engineer • Coding • hard

Given an array of integers representing the execution times of ML training jobs and an integer K representing the number of available GPUs, partition the jobs to minimize the maximum execution time on any single GPU. Jobs must be scheduled in contiguous subarrays.

#Binary Search #Greedy Algorithms #Dynamic Programming

Practice

Machine Learning Engineer • System Design • hard

Design an autocomplete/typeahead system for Google Docs using a neural language model. The system must run within strict latency constraints (<50ms). How do you optimize the model and serving infrastructure?

#Low Latency Serving #Model Quantization #Sequence-to-Sequence #Edge ML

Practice

Machine Learning Engineer • System Design • hard

Design the recommendation system for YouTube Shorts. Specifically, how would you handle the cold-start problem for new creators and optimize for real-time engagement metrics like watch time and swipe-aways?

#Recommendation Systems #Two-Tower Models #Cold Start #Real-time Streaming

Practice

Machine Learning Engineer • System Design • hard

Design a system to predict Ad Click-Through Rate (CTR) for Google Search. How do you handle categorical features with massive cardinality, and how do you update the model with fresh data throughout the day?

#CTR Prediction #Feature Engineering #Continuous Training #Embeddings

Practice

Machine Learning Engineer • System Design • medium

Design a system to detect policy-violating images (e.g., hate speech, extreme violence) uploaded to Google Drive. The system must process millions of images per minute with extreme precision to avoid false positives on user data.

#Computer Vision #High Throughput #Active Learning #Anomaly Detection

Practice

Machine Learning Engineer • Technical • medium

Explain the mathematical difference between Layer Normalization and Batch Normalization. Why is Layer Normalization almost exclusively used in Transformer architectures instead of Batch Normalization?

#Normalization #Transformers #Mathematics

Practice

Machine Learning Engineer • Technical • hard

Explain how you would implement KV-caching in a Transformer model during autoregressive inference. What are the memory bottlenecks, and how do techniques like PagedAttention address them?

#Transformers #LLM Inference #Memory Optimization #Attention Mechanisms

Practice

Machine Learning Engineer • Technical • medium

How would you evaluate the quality of a Retrieval-Augmented Generation (RAG) system built for Google Cloud enterprise search? What specific metrics would you use for the retrieval component vs. the generation component?

#RAG #LLM Evaluation #Information Retrieval #Metrics

Practice

Machine Learning Engineer • Technical • medium

You are training a multimodal model (text and image) using a contrastive loss similar to CLIP. You notice the text loss converges much faster than the image loss, leading to poor alignment. How do you diagnose and fix this?

#Multimodal ML #Loss Optimization #Contrastive Learning #Debugging

Practice

Machine Learning Engineer • Technical • hard

How does Distributed Data Parallel (DDP) differ from Fully Sharded Data Parallel (FSDP) or ZeRO optimization when training large language models? When would you choose one over the other?

#Distributed Training #Model Parallelism #Data Parallelism #Memory Management

Practice

ML Engineer • Behavioral • medium

Tell me about a disagreement you had with a researcher. How did you resolve it?

#Communication

Practice

ML Engineer • Behavioral • easy

How do you keep up with the rapidly evolving ML landscape?

#Continuous Learning

Practice

ML Engineer • Behavioral • hard

Tell me about a time an ML model caused an unexpected real-world impact.

#Responsibility #AI Safety

Practice

ML Engineer • Behavioral • medium

Describe how you collaborated with data scientists to productionize their research code.

#Research to Production

Practice

ML Engineer • Behavioral • hard

Tell me about a time you had to optimize a model for latency without sacrificing too much accuracy.

#Latency #Accuracy

Practice

ML Engineer • Behavioral • medium

Describe a model you deployed to production. What were the biggest challenges?

#Deployment #Challenges

Practice

ML Engineer • Behavioral • medium

How do you decide when a model is 'good enough' to ship?

#Quality #Judgment

Practice

ML Engineer • Behavioral • hard

Describe a time you had to re-architecture a system because the original ML approach didn't scale.

#Scalability

Practice

ML Engineer • Coding • hard

Write a custom PyTorch Dataset and DataLoader for irregular time series data.

#PyTorch #DataLoader

Practice

ML Engineer • Coding • hard

Implement logistic regression with gradient descent in NumPy.

#Logistic Regression #NumPy

Practice

ML Engineer • Coding • hard

Implement a K-means clustering algorithm from scratch in Python.

#K-Means #Clustering

Practice

ML Engineer • Coding • hard

How would you write a batched inference pipeline using Python and Triton server?

#Triton #Batching

Practice

ML Engineer • Coding • medium

Implement a sliding window approach to detect anomalies in a time series.

#Anomaly Detection #Time Series

Practice

ML Engineer • System Design • hard

Design a system to retrain models automatically when performance degrades.

#Retraining #Automation

Practice

ML Engineer • System Design • hard

Design YouTube's video recommendation system end to end.

#Recommendations #Ranking

Practice

ML Engineer • System Design • hard

How would you build a personalized ad targeting system?

#Targeting #ML Systems

Practice

ML Engineer • System Design • hard

Design a training and serving architecture for a large language model at scale.

#Infrastructure #Scale

Practice

ML Engineer • System Design • hard

Design a search ranking system for an e-commerce platform.

#Ranking #Relevance

Practice

ML Engineer • System Design • hard

What is a feature store? Design one from scratch.

#Feature Engineering #MLOps

Practice

ML Engineer • System Design • hard

Design a CI/CD pipeline for ML models.

#CI/CD #Deployment

Practice

ML Engineer • System Design • hard

Design a real-time content moderation system.

#NLP #Real-Time

Practice

ML Engineer • System Design • hard

How would you serve a model that needs to respond in under 10ms?

#Low Latency #Serving

Practice

ML Engineer • Technical • hard

How do you detect data drift vs model drift? How do you respond to each?

#Drift #Production

Practice

ML Engineer • Technical • medium

How would you deploy a model with Vertex AI Predictions?

#Vertex AI #GCP

Practice

ML Engineer • Technical • medium

What is model ensembling? When does it help, and when does it hurt?

#Ensembling #Performance

Practice

ML Engineer • Technical • medium

Explain vector databases. What are FAISS, Pinecone, and Weaviate?

#Vector DB #Embeddings

Practice

ML Engineer • Technical • hard

How would you evaluate an LLM for a production use case?

#Evaluation #Benchmarking

Practice

ML Engineer • Technical • medium

What is Vertex AI? How does it compare to SageMaker?

#Vertex AI #SageMaker

Practice

ML Engineer • Technical • hard

What is quantization in neural networks? How does it reduce inference cost?

#Quantization #Inference

Practice

ML Engineer • Technical • hard

Explain knowledge distillation. When would you use it?

#Distillation #Compression

Practice

ML Engineer • Technical • hard

What is the difference between model parallelism and data parallelism in distributed training?

#Parallelism #Training

Practice

ML Engineer • Technical • medium

How do you version ML models and datasets? What tools do you use?

#Versioning #DVC #MLflow

Practice

ML Engineer • Technical • hard

Explain blue-green deployment vs canary deployment for ML models.

#Blue-Green #Canary

Practice

ML Engineer • Technical • medium

What is shadow mode deployment in ML?

#Shadow Mode #A/B Testing

Practice

ML Engineer • Technical • hard

What is RAG (Retrieval-Augmented Generation)? Describe its architecture.

#RAG #Vector Search

Practice

ML Engineer • Technical • hard

What is LoRA (Low-Rank Adaptation)? How does it reduce fine-tuning costs?

#LoRA #Fine-Tuning

Practice

ML Engineer • Technical • hard

Explain the RLHF (Reinforcement Learning from Human Feedback) training approach.

#RLHF #Fine-Tuning

Practice

ML Engineer • Technical • medium

How do you profile and debug a slow training run?

#Profiling #Debugging

Practice

ML Engineer • Technical • medium

What are the differences between PyTorch and TensorFlow for production?

#PyTorch #TensorFlow

Practice

ML Engineer • Technical • hard

Explain mixed precision training (FP16/BF16). What are the risks?

#Mixed Precision #Performance

Practice

ML Engineer • Technical • hard

How do you optimize GPU utilization during training?

#GPU #Performance

Practice

ML Engineer • Technical • medium

What is Kubernetes? How is it used for ML model serving?

#Kubernetes #Serving

Practice

ML Engineer • Technical • medium

Explain model serialization formats: ONNX, TorchScript, SavedModel.

#ONNX #Serialization

Practice

ML Engineer • Technical • hard

Explain how Google's Two-Tower model works for recommendations.

#Two-Tower #Embeddings

Practice

ML Engineer • Technical • hard

How does Google's TensorFlow Extended (TFX) pipeline work?

#TFX #Pipelines

Practice

ML Engineer • Technical • easy

What is the difference between a data scientist and an ML engineer?

#Roles #MLOps

Practice

ML Engineer • Technical • medium

Explain the model training pipeline from raw data to deployment.

#Pipeline #Training

Practice

ML Engineer • Technical • medium

What is the difference between online learning and offline learning?

#Online Learning #Batch Learning

Practice

ML Engineer • Technical • medium

How do you handle missing data in ML model features?

#Imputation #Missing Data

Practice

ML Engineer • Technical • medium

Explain gradient descent variants: batch, stochastic, and mini-batch.

#Gradient Descent #Optimization

Practice

ML Engineer • Technical • medium

What are learning rate schedulers and why are they important?

#Learning Rate #Training

Practice

ML Engineer • Technical • hard

Explain the attention mechanism in transformers with mathematical detail.

#Attention #Transformers

Practice

Product Manager • Behavioral • medium

Tell me about a time you had to align conflicting stakeholders across engineering and design on a tight deadline.

#Stakeholder Management #Conflict Resolution #Cross-functional Collaboration

Practice

Product Manager • Behavioral • hard

Describe a situation where you had to convince an engineering team to build a feature they strongly disagreed with.

#Influence without Authority #Engineering Collaboration #Communication

Practice

Product Manager • Behavioral • medium

Tell me about a product or feature you launched that failed. What metrics indicated it failed, and what did you learn?

#Post-mortem #Resilience #Data-driven

Practice

Product Manager • Coding • easy

Write a SQL query to find the top 3 most watched YouTube video categories in the last 30 days, given a 'views' table and a 'videos' table.

#SQL #Data Analysis #YouTube

Practice

Product Manager • System Design • hard

Explain how Google Search autocomplete works at a high level and how you would scale it for a newly supported language.

#Google Search #Latency #Scalability #Data Structures

Practice

Product Manager • System Design • hard

Design the backend architecture for a real-time collaborative editing feature in Google Docs.

#Google Docs #Concurrency #Distributed Systems

Practice

Product Manager • Technical • medium

You are the PM for Google Chrome. You have a proposed feature that increases page load speed by 5% but increases memory usage by 15%. Do you launch it?

#Google Chrome #Trade-offs #Performance Metrics

Practice

Product Manager • Technical • medium

Design a product for Google Maps that helps users find parking using AI.

#Google Maps #Artificial Intelligence #User Experience

Practice

Product Manager • Technical • hard

Microsoft is aggressively integrating ChatGPT into Bing. What should be Google Search's strategic response over the next 2 years?

#Google Search #Competitive Analysis #Generative AI #Gemini

Practice

Product Manager • Technical • medium

YouTube Shorts engagement has dropped by 10% week-over-week. How do you investigate and resolve this?

#YouTube Shorts #Root Cause Analysis #Data Analytics

Practice

Product Manager • Technical • hard

How would you integrate Gemini (Google's LLM) into Google Workspace (Docs/Sheets) specifically for enterprise B2B users?

#Generative AI #Google Workspace #B2B Enterprise

Practice

Product Manager • Technical • medium

Estimate the total bandwidth consumed by Google Photos backups globally in a single day.

#Google Photos #Fermi Problem #Data Storage

Practice

Product Manager • Technical • easy

Design a Google Nest smart display device specifically for the elderly.

#Google Nest #Accessibility #Hardware PM

Practice

Product Manager • Technical • medium

How would you monetize Google Maps further without showing intrusive ads on the core map interface?

#Google Maps #Monetization #B2B APIs

Practice

Product Manager • Technical • hard

Should Google spin off YouTube into a completely separate company? Walk me through your strategic reasoning.

#YouTube #Business Strategy #Market Dynamics

Practice

Software Engineer • Behavioral • medium

Tell me about a time you pushed back on a product manager's feature request because you believed it would negatively impact system latency or reliability. How did you resolve the conflict?

#Communication #Googlyness #Prioritization #Conflict Resolution

Practice

Software Engineer • Behavioral • medium

Describe a time when you had to lead a project across multiple timezones or distributed teams. How did you ensure alignment and handle communication breakdowns?

#Cross-functional Collaboration #Leadership #Project Management

Practice

Software Engineer • Behavioral • easy

Tell me about a time you discovered a critical bug or security vulnerability in a production system. What was your immediate action, and how did you ensure it wouldn't happen again?

#Incident Management #Post-mortem #Ownership

Practice

Software Engineer • Behavioral • medium

Tell me about a time you had to pivot your technical approach completely because of changing business requirements from product managers. How did you manage the transition and team morale?

#Adaptability #Conflict Resolution #Team Dynamics

Practice

Software Engineer • Behavioral • medium

Describe a situation where you had to navigate a highly ambiguous project with no clear technical direction. How did you define the milestones and deliver the solution?

#Ambiguity #Project Management #Googlyness

Practice

Software Engineer • Behavioral • medium

Tell me about a time you discovered a significant flaw in a system's architecture right before a major launch. How did you balance the need to ship on time with the need to fix the technical debt?

#Googlyness #Decision Making #Risk Management #Communication

Practice

Software Engineer • Coding • medium

You are given a stream of Google Ads click events. Implement a sliding window counter that returns the number of clicks in the exact last 5 minutes. The stream is highly concurrent.

#Concurrency #Sliding Window #Queues #Data Streams

Practice

Software Engineer • Coding • medium

Given a list of Google Calendar events for 'n' users, where each event consists of a start and end time, and a required meeting duration 'k', find all available time slots where all 'n' users can attend a meeting.

#Intervals #Two Pointers #Sorting

Practice

Software Engineer • Coding • medium

Implement a system for Google Calendar that takes in a list of N users' daily schedules (lists of busy intervals) and their working hours, and returns all available time slots of duration T where all N users can meet.

#Intervals #Sorting #Two Pointers #Arrays

Practice

Software Engineer • Coding • hard

Given a list of daily budgets and expected returns for various Google Ads campaigns, write an algorithm to allocate a total budget B to maximize the overall return. You can allocate partial budgets to campaigns, but returns diminish non-linearly.

#Dynamic Programming #Greedy Algorithms #Optimization #Math

Practice

Software Engineer • Coding • hard

Design a data structure for Google Search Autocomplete that supports adding new search queries, updating the frequency of a query, and retrieving the top 'k' most frequent queries that start with a given prefix in real-time.

#Trie #Heap #Hash Map #Design

Practice

Software Engineer • Coding • hard

You are given a map represented as a weighted directed graph. You need to route an electric vehicle from point A to point B. The EV has a maximum battery capacity, and certain nodes have charging stations. Find the shortest path such that the EV never runs out of battery.

#Graphs #Dijkstra's Algorithm #State Space Search

Practice

Software Engineer • Coding • medium

You are given a map represented as a 2D grid where 0 is a road, 1 is a building, and 2 is an EV charging station. Given a starting position, find the shortest path to an EV charging station. Follow-up: How would you optimize this if we have millions of queries per second for different starting positions on a static map?

#Graph Theory #BFS #Dynamic Programming #Caching

Practice

Software Engineer • Coding • medium

Implement a thread-safe LRU (Least Recently Used) cache. This cache will be used in a high-throughput microservice. Explain how you would minimize lock contention.

#Concurrency #Data Structures #Linked List #Hash Map

Practice

Software Engineer • Coding • hard

You are building a feature for Google Docs. Given a string representing a document and a list of operations (insert, delete, replace at specific indices), apply the operations efficiently. How do you handle overlapping operations?

#String Manipulation #Operational Transformation #Design

Practice

Software Engineer • Coding • medium

Given a stream of user watch events on YouTube (user_id, video_id, timestamp, duration_watched), write a function to find the longest contiguous sequence of videos a user watched where they completed at least 90% of each video.

#Sliding Window #Hash Map #Stream Processing

Practice

Software Engineer • Coding • hard

Design a data structure for Google Search autocomplete. It must support inserting a string with a frequency, and querying the top K most frequent strings that start with a given prefix. Optimize for query latency.

#Trie #Priority Queue #Design #String Manipulation

Practice

Software Engineer • Coding • medium

Implement a function to evaluate a mathematical expression given as a string (e.g., '3 + 5 / (2 - 1) * 4'). The expression can contain parentheses, and you must follow standard order of operations. This is used in Google Search's calculator widget.

#Stacks #String Parsing #Math

Practice

Software Engineer • Coding • medium

You are given an array of integers representing the CPU load of a Google server cluster over time. Find the maximum contiguous subarray sum, but you are allowed to delete exactly one element to maximize the sum.

#Dynamic Programming #Arrays #Kadane's Algorithm

Practice

Software Engineer • Coding • hard

Given a 2D grid representing a Google Maps satellite image where '1' is land and '0' is water, find the minimum number of days to connect two disconnected islands. You can change one '0' to '1' per day.

#BFS #DFS #Matrix

Practice

Software Engineer • System Design • medium

Design a distributed rate limiter for Google Cloud API Gateway that can handle millions of requests per second with minimal latency overhead.

#Redis #Token Bucket #Distributed Systems #Hashing

Practice

Software Engineer • System Design • hard

Design the real-time view counter for YouTube live streams. The system must handle massive spikes in traffic (e.g., a Super Bowl stream) and provide eventually consistent view counts to the UI with sub-second latency.

#Stream Processing #Event Sourcing #Scalability #Data Aggregation

Practice

Software Engineer • System Design • hard

Design Google Photos' auto-backup feature for mobile devices. How do you handle intermittent network connectivity, deduplication, and efficient storage?

#Blob Storage #Checksum #Mobile Sync #Resumable Uploads

Practice

Software Engineer • System Design • hard

Design Google Search's ranking pipeline.

#Ranking #Scale

Practice

Software Engineer • System Design • hard

Design the backend for Google Docs collaborative editing. How do you handle concurrent edits from multiple users offline and online to ensure eventual consistency?

#Operational Transformation #CRDTs #WebSockets #Concurrency

Practice

Software Engineer • System Design • hard

Design a distributed, highly available rate limiter for Google Cloud APIs. It needs to support millions of requests per second, enforce limits per customer per API, and add minimal latency to the critical path.

#Distributed Caching #Token Bucket #Redis #High Availability

Practice

Software Engineer • System Design • hard

How would you design a distributed key-value store like Bigtable?

#Key-Value Store

Practice

Software Engineer • System Design • hard

Design the block-level file synchronization mechanism for Google Drive. How do you handle concurrent edits offline, minimize bandwidth for large files, and resolve conflicts when the client reconnects?

#Distributed Systems #Data Synchronization #Concurrency #Network Optimization

Practice

Software Engineer • System Design • hard

Design the video recommendation feed for YouTube. Focus on how you would fetch, rank, and serve the recommendations at scale within a 200ms latency budget.

#Machine Learning Infra #Caching #Microservices #Recommendation Systems

Practice

Software Engineer • System Design • hard

Design a distributed web crawler for Google Search. How do you handle DNS resolution bottlenecks, avoid crawler traps, prioritize high-quality domains, and ensure you don't DDoS the target servers?

#Distributed Systems #Graph Traversal #Politeness Policies #Queueing

Practice

Software Engineer • Technical • medium

Implement a thread-safe LRU cache in your language of choice. It must support get() and put() in O(1) time, and handle concurrent access from multiple threads without race conditions or deadlocks.

#Concurrency #Hash Map #Doubly Linked List #Mutex

Practice

Software Engineer • Technical • medium

Given two tables: `search_logs` (query_id, user_id, query_string, timestamp, region) and `clicks` (query_id, url_clicked, rank_position), write an optimized SQL query to find the top 3 queries with the highest click-through rate in the 'US' region over the last 7 days, partitioned by day.

#Window Functions #Joins #Aggregations #Performance Optimization

Practice

Software Engineer • Technical • hard

What is MapReduce? How does it work at Google's scale?

#MapReduce

Practice

Google

The Interview Loop

Recruiter Screen (30 min)

Technical Loop (3-4 Rounds)

Interview Question Bank

Tell me about an AI project where you had to balance innovation with reliability.

How do you handle stakeholder uncertainty around AI capabilities and limitations?

Tell me about a time you optimized an LLM application for cost or latency.

Describe a time you had to choose between using an AI model and a simpler rule-based system.

How do you stay current with the fast-moving AI/ML research landscape?

Tell me about a time an AI system you built produced unexpected or harmful outputs.

Describe an AI product you built from scratch. What were the key technical decisions?

Describe a situation where you had to debug a hard-to-reproduce AI model failure.

Implement a semantic chunking strategy for long documents.

Write a Python class to manage conversation history for a multi-turn chatbot.

Implement a simple RAG pipeline using Python, LangChain, and FAISS.

Write a retry mechanism with exponential backoff for LLM API calls.

How would you architect an AI platform that supports 1000 concurrent LLM requests?

Design an AI-powered customer support chatbot for an e-commerce platform.

Design a document question-answering system using RAG.

Design an AI code review system that integrates with GitHub PRs.

How would you build a multi-modal AI system that processes both text and images?

Design a real-time AI safety filter for user-generated content.

Design an AI agent system that can autonomously browse the web and complete tasks.

How do you choose the right embedding model for a domain-specific search task?

Explain positional encoding in transformers. What are the differences between absolute and rotary position embeddings?

What is hallucination in LLMs? How do you detect and mitigate it?

Explain the difference between autoregressive and masked language modeling.

What is a mixture of experts (MoE) architecture? How does it scale?

What is Constitutional AI? How does Anthropic use it?

How do you red-team an AI system?

What are guardrails in LLM applications? How do they work?

How do you integrate OpenAI API or Gemini API into a production application?

What is LangChain? What are its key components (Chains, Agents, Tools)?

What is streaming response from an LLM API? How do you implement it in a web app?

Explain structured output generation from LLMs (JSON mode, Instructor library).

Explain how vector similarity search works. What are HNSW and IVF indices?

Compare vector databases: Pinecone, Weaviate, Qdrant, and pgvector.

What is semantic search? How does it differ from keyword-based search?

Explain the difference between dense and sparse retrieval in RAG.

How do you evaluate retrieval quality in a RAG system?

How do you evaluate the quality of an LLM-generated response?

What is AI alignment? What are the key safety concerns with large-scale AI deployment?

Explain the concept of AI bias. How do you detect and mitigate it in production?

How do you manage LLM API rate limits and costs in production?

Explain function calling / tool use in LLMs. How do you implement it?

Explain the difference between GPT, BERT, and T5 architectures.

What is prompt engineering? What are few-shot, zero-shot, and chain-of-thought prompting?

Explain how RLHF (Reinforcement Learning from Human Feedback) improves LLMs.

What is RAG (Retrieval-Augmented Generation)? When would you use it over fine-tuning?

Explain the difference between fine-tuning and in-context learning.

What is token context window? How do you handle documents longer than the context limit?

Tell me about a time you significantly reduced cloud infrastructure costs.

Describe a situation where a critical production system went down, and there was no runbook. How did you handle it?

Tell me about a time you had to work with a difficult stakeholder or team member who strongly disagreed with your technical approach. How did you resolve it?

Tell me about a time you had to push back on a customer's architectural choice because you knew it would lead to scalability issues down the line.

Describe your experience with incident post-mortems. What do you include?

How do you communicate a complex cloud architecture to non-technical stakeholders?

Tell me about a time you improved the reliability of a cloud-based data system.

Describe a situation where you had to choose between two cloud architectures. How did you decide?

Tell me about a major cloud outage you experienced. How did you respond?

Describe a time you migrated a critical workload to the cloud with zero downtime.

How do you stay updated with new cloud services and features?

Write a function to validate if a given string is a valid IPv4 address, and then extend it to check if it belongs to a specific CIDR block.

Given a list of log entries with timestamps and error codes, write a function to find the top 3 most frequent error codes within a sliding window of 5 minutes.

Write a Python script to find and delete all unattached persistent disks in a GCP project that are older than 30 days to save costs.

How would you set up a streaming data pipeline on GCP using Pub/Sub and Dataflow?

Design a data lake on AWS using S3, Glue, and Athena.

Design a real-time streaming data pipeline on GCP to ingest, process, and analyze millions of IoT sensor events per second.

Design a highly available, globally distributed web application on GCP that handles sudden, massive spikes in traffic (e.g., a viral news site).

A customer wants to migrate a monolithic on-premise application backed by an Oracle database to GCP. Walk me through your migration strategy.

How do you implement disaster recovery for a cloud data warehouse?

How would you architect a data platform that reduces spend by 40% without impacting performance?

What is the shared responsibility model in cloud security?

What is a VPC (Virtual Private Cloud)? How do you design a secure VPC architecture?

Explain the difference between regions, availability zones, and edge locations.

How does auto-scaling work? What are the different scaling strategies?

What is a cloud-native application? How does it differ from a lifted-and-shifted one?

Explain multi-cloud vs hybrid cloud architectures and their tradeoffs.

Explain Kubernetes architecture: control plane, nodes, pods, and services.