SQL is the language of databases, and AI is very good at writing it. Analysts, marketers and founders can now pull their own reports without waiting on engineering.
Share the schema first
AI can't guess your table and column names. Paste the table structure (not the data):
Tables:
orders(id, customer_id, total, status, created_at)
customers(id, name, country, signup_date)
Write a MySQL query that shows total revenue per country for
completed orders in the last 90 days, highest first.Ask for an explanation
Request a line-by-line explanation of the query. Over time you'll learn SQL just by reading these.
Common requests AI handles well
- Joins across several tables
- Grouping and aggregations (sums, counts, averages)
- Date ranges and time buckets (per week, per month)
- Window functions like running totals and rankings
- Finding duplicates or missing records
Check the results
- Test on a small date range and compare with a number you know.
- Watch for joins that duplicate rows and inflate totals.
- Run read-only queries; never let AI-written UPDATE or DELETE statements run on production without review and a backup.