Data Engineering / Exactly Once Interview Questions 2026

Published July 22, 2026 · Updated July 22, 2026

17 Data Engineering / Exactly Once interview questions from General, Amazon, Deloitte and more. Real questions from Technical, System Design, and HR rounds.

17 questions
Companies: General, Amazon, Deloitte

All Data Engineering / Exactly Once Questions

General Medium Screening round 3-7 years

What is a data lakehouse and how does it differ from a data lake + data warehouse architecture?

↑ 72 upvotes · 10 engineers asked this · SDE2
General Hard System Design round 5-9 years

Design a pipeline to detect fraudulent transactions in near-real-time. Transactions arrive at 10,000/second. Your ML model takes 5ms to evaluate one transaction.

↑ 71 upvotes · 9 engineers asked this · Staff
General Medium Screening round 2-5 years

Explain the difference between batch processing and stream processing. Can you use Spark for both?

↑ 70 upvotes · 8 engineers asked this · SDE2
General Hard Scenario round 3-7 years

Your data warehouse query returns the wrong totals for a report. The same query on raw data shows different numbers. How do you investigate this?

↑ 69 upvotes · 7 engineers asked this · SDE2
General Hard Screening round 4-8 years

Explain the CAP theorem in the context of choosing between Apache Kafka and a traditional database for event sourcing.

↑ 68 upvotes · 6 engineers asked this · SDE2
General Hard Scenario round 3-7 years

You have a Spark job that runs fine on 1GB data but OOMs on 100GB. Your cluster has 20 nodes with 16GB each. How do you diagnose and fix this?

↑ 67 upvotes · 5 engineers asked this · SDE2
General Medium Screening round 3-7 years

Explain the difference between a star schema and a snowflake schema in a data warehouse. Which is better for query performance?

↑ 66 upvotes · 4 engineers asked this · SDE2
General Hard System Design round 4-8 years

Your data pipeline reads from Kafka, transforms data, and writes to a data warehouse. On failure, it must resume from exactly where it left off, without duplicates or missing records. How?

↑ 65 upvotes · 3 engineers asked this · SDE2
Amazon Hard Screening round 4-8 years

Explain the CAP theorem and how NoSQL databases make trade-offs around it.

↑ 50 upvotes · 48 engineers asked this · SDE2
General Hard Technical round 4-7 years

You have a Spark job that processes 1TB of data. It runs fine for 900GB but fails with OOM on the remaining 100GB. Why?

↑ 49 upvotes · 47 engineers asked this · SDE2
General Medium Screening round 3-6 years

What is a star schema vs snowflake schema in data modeling? Trade-offs?

↑ 48 upvotes · 46 engineers asked this · SDE2
General Hard Screening round 4-7 years

Explain partitioning strategies in Parquet files on S3. How does column pruning and predicate pushdown reduce query cost?

↑ 47 upvotes · 45 engineers asked this · SDE2
General Hard Screening round 3-7 years

What are slowly changing dimensions (SCD) in data warehousing? Explain Type 1, Type 2, and Type 3.

↑ 46 upvotes · 44 engineers asked this · SDE2
General Hard Screening round 4-7 years

Explain the difference between batch and streaming data processing. When is Flink better than Spark for streaming?

↑ 45 upvotes · 43 engineers asked this · SDE2
General Medium Screening round 2-5 years

What is Apache Airflow and how does it differ from a cron job for orchestrating data pipelines?

↑ 44 upvotes · 42 engineers asked this · SDE2
General Medium Screening round 3-6 years

Explain the differences between a Data Warehouse, Data Lake, and Data Lakehouse architecture.

↑ 43 upvotes · 41 engineers asked this · SDE2
Deloitte Medium Technical round 3-6 yearsData Engineering

A nightly ETL pipeline occasionally fails midway, leaving the target table with partial data for that day. How would you redesign the pipeline to make a failed run safe to simply re-run?

↑ 21 upvotes · 9 engineers asked this · Data Engineer

Practice these questions with AI feedback

Get instant grading on your answers, identify your weak areas, and generate a personalised 14-day study plan — all free.

Build my study plan →

Same topic at specific companies

General Data Engineering / Exactly Once Amazon Data Engineering / Exactly Once Deloitte Data Engineering / Exactly Once

Related topics

DSA / Arrays System Design DSA / Trees DSA / Graphs Behavioral Testing Domain DevOps API Testing DSA / Hashmaps GCP / Real-time Data Java / Pub-Sub React / Performance Node.js / Memory SQL / Query Optimization Python / Async AWS / Lambda Kafka / Consumer Microservices / Saga Pattern TypeScript / Type System MongoDB / Indexing Security / SQL Injection Performance / Latency Tail Redis / Cache Stampede Elasticsearch / Cluster Health GraphQL / Security Snowflake / Architecture PySpark / Core Concepts Airflow / Scheduling Web / WebAssembly Debugging / Production JavaScript / V8 Engine

Frequently asked questions

What is a data lakehouse and how does it differ from a data lake + data warehouse architecture?
What is a data lakehouse and how does it differ from a data lake + data warehouse architecture?
Design a pipeline to detect fraudulent transactions in near-real-time. Transactions arrive at 10,000/second. Your ML model takes 5ms to evaluate one transaction.
Design a pipeline to detect fraudulent transactions in near-real-time. Transactions arrive at 10,000/second. Your ML model takes 5ms to evaluate one transaction.
Explain the difference between batch processing and stream processing. Can you use Spark for both?
Explain the difference between batch processing and stream processing. Can you use Spark for both?
Your data warehouse query returns the wrong totals for a report. The same query on raw data shows different numbers. How do you investigate this?
Your data warehouse query returns the wrong totals for a report. The same query on raw data shows different numbers. How do you investigate this?
Explain the CAP theorem in the context of choosing between Apache Kafka and a traditional database for event sourcing.
Explain the CAP theorem in the context of choosing between Apache Kafka and a traditional database for event sourcing.