Interview practice

Data Engineer

2–5 years · Mid — SQL · Python · Spark · Airflow · Warehousing · ETL / ELT

technical Data Quality 1 of 6

A daily data quality check reveals that 5% of your records are missing key attributes. What systematic approach do you take to identify and resolve this issue?

What a strong answer covers

Analyzes the transformation logic to find where data might be dropped; checks source data for completeness; implements logging to track record counts through each pipeline stage; creates a temporary table to isolate missing records; reviews upstream processes for changes that might affect data completeness.

The follow-up you would get

After identifying the source of the missing data, what steps do you take to implement a fix?

Could you have answered that?

technical Data Pipeline 2 of 6

An ETL job fails every night at 3am with a timeout error after processing 10 million records. How do you diagnose the issue, and what are your immediate next steps?

What a strong answer covers

Checks the job logs for error messages; identifies the stage at which the timeout occurs; reviews data volume and processing time for previous runs; inspects resource allocation and concurrency settings; considers whether data skew is affecting performance.

The follow-up you would get

The timeout seems related to a specific transformation. What transformations would you look at first?

Could you have answered that?

scenario Data Quality 3 of 6

A downstream team reports that the data they received is incorrect, but your pipeline shows successful execution. What do you investigate first?

What a strong answer covers

Checks the integrity of the data source against the expected schema; verifies transformation logic for anomalies; reviews logs for warnings or errors during processing; confirms the data lineage to identify where the issue may have arisen; engages with the downstream team for specific examples of discrepancies.

The follow-up you would get

You discover a transformation step that misapplied business rules. How do you prevent this from happening in future iterations?

Could you have answered that?

ownership A migration you led 4 of 6

Can you describe a migration project you led, including any challenges faced and how you overcame them?

What a strong answer covers

Names the system migrated and rationale for the migration; discusses specific technical challenges encountered; describes the strategies used to mitigate risks; details the testing process to ensure data integrity; reflects on the overall impact of the migration on the team or business.

The follow-up you would get

What unexpected issues arose during the migration?

Could you have answered that?

scenario Error Handling 5 of 6

Your ETL job encounters an unexpected data format and aborts. How do you handle this situation to ensure data continuity?

What a strong answer covers

Sets up a robust error-handling mechanism to log the error and alert the team; implements a fallback procedure to skip erroneous records while continuing processing; reviews the data validation rules to catch format issues early; schedules a review session to discuss the data source integrity; considers implementing a dead-letter queue for persistent errors.

The follow-up you would get

After implementing skipping logic, you find out that the skipped records contain critical information. How do you address this?

Could you have answered that?

scenario Performance Tuning 6 of 6

A business user complains that the reports generated from your data warehouse are taking too long to load. What steps do you take to resolve the issue?

What a strong answer covers

Analyses the query execution plans for bottlenecks; checks for inefficient joins or missing indexes in the queries; evaluates the hardware resources and database configuration; considers caching strategies for frequently accessed data; consults with stakeholders to understand usage patterns.

The follow-up you would get

You identify a query that is poorly optimized. How do you approach rewriting it to improve performance?

Could you have answered that?

That is this set.

Load six more — you will get questions you have not seen first.

Give me six more →

Say your answer out loud before revealing. Reading the answer first feels like learning and is not.