SQL interview preparation should cover filtering, joins, grouping, keys, NULL values, and basic window functions. A strong answer explains the concept, demonstrates it with a query, and checks the result. Learn how to reason about data rather than memorising answers without understanding what the query returns.<\/p>\n\n\n\n
The examples below use PostgreSQL-compatible syntax. Other databases may differ. Ask which database and version the interviewer expects before using dialect-specific features.<\/p>\n\n\n\n
What is SQL?<\/strong><\/p>\n\n\n\n SQL stands for Structured Query Language. It is used to work with relational databases, including retrieving and changing data and defining database structures.<\/p>\n\n\n\n A concise interview answer can connect the definition to a task: \u201cI use SQL to filter records, combine related tables, and calculate summaries.\u201d Then provide an example from a project you actually completed.<\/p>\n\n\n\n What is the difference between WHERE and HAVING?<\/strong><\/p>\n\n\n\n WHERE filters input rows before grouping. HAVING filters the resulting groups. Suppose an orders table contains customer_id, amount, and status:<\/p>\n\n\n\n SELECT customer_id, SUM(amount) AS total This query first selects paid orders, then calculates each customer’s total and keeps totals above 10,000. PostgreSQL’s aggregate tutorial explains the distinction.<\/p>\n\n\n\n In an interview, say what one output row represents: one qualifying customer. That small explanation helps demonstrate that you understand the result.<\/p>\n\n\n\n How do INNER JOIN and LEFT JOIN differ?<\/strong><\/p>\n\n\n\n INNER JOIN returns matching row combinations. LEFT JOIN also preserves rows from the left table when no match exists on the right, filling the right-side columns with NULL.<\/p>\n\n\n\n For example, join customers to orders with LEFT JOIN when you need to retain customers who have not ordered. Check the relationship carefully: one customer with three orders can create three result rows.<\/p>\n\n\n\n The PostgreSQL join tutorial demonstrates these behaviors. Explain how you would avoid counting joined order rows as unique customers.<\/p>\n\n\n\n
FROM orders
WHERE status = ‘paid’
GROUP BY customer_id
HAVING SUM(amount) > 10000;<\/p>\n\n\n\n