Query performance matters in data engineering because slow SQL can delay reporting, overload shared databases, and increase pipeline runtimes. In PostgreSQL, optimization is not just about writing shorter SQL; it is about understanding how the planner executes the query.
You are asked to explain how you optimize a slow SQL query in PostgreSQL. Walk through how you would diagnose the bottleneck, what you would look for in an execution plan, when you would add or change indexes, and when you would rewrite the query instead of relying on indexing alone.
The interviewer expects a practical, medium-depth answer. You should cover a structured debugging approach, common performance issues such as full table scans and expensive joins, and the trade-offs of indexing, filtering, and aggregation. Mention PostgreSQL-specific tools like EXPLAIN or EXPLAIN ANALYZE and show that you can connect query design to database performance.