Loading...
Loading...
Questions covering data pipeline design, ETL, warehousing, and distributed computing for data engineering roles.
7 questions with expert tips
Practice with AI →Explain the difference between a star schema and a snowflake schema. When would you prefer each?
Expert Tip
Star: denormalized, faster queries, more storage. Snowflake: normalized, slower queries, less storage. Use star for analytics.
Write a SQL query to find the top 3 products by revenue in each category for the last 30 days.
Expert Tip
Use ROW_NUMBER() or RANK() with PARTITION BY category ORDER BY revenue DESC, then filter WHERE rn <= 3.
How would you design a data pipeline that ingests 1 billion rows per day from IoT sensors?
Expert Tip
Kafka for streaming ingestion, Spark Structured Streaming for processing, Parquet on S3/GCS, partition by date/device.
Your ETL pipeline failed at step 3 of 7 after processing 40% of records. How do you handle idempotency?
Expert Tip
Checkpointing, upserts (INSERT OR REPLACE), partition pruning to reprocess only affected partitions, idempotent writes.
How do you test data quality in a dbt project? What tests would you set on a critical user_id column?
Expert Tip
dbt test: not_null, unique, accepted_values, referential_integrity. Custom tests for freshness, row count thresholds.
A Spark job is running for 4 hours and analysts are complaining. How do you diagnose and fix it?
Expert Tip
Check Spark UI for skew, shuffles, spill. Solutions: repartition, broadcast joins for small tables, cache intermediate results.
Explain the medallion architecture (Bronze/Silver/Gold). How does it improve data governance?
Expert Tip
Bronze: raw ingestion. Silver: cleansed, validated. Gold: business-level aggregates. Separation of concerns, reprocessing.
Answer these questions in a simulated interview and get real-time AI feedback on your responses, pacing, and depth.
Start Practice Interview