How does iDBQuery write the SQL for me?
iDBQuery writes SQL for you by combining your plain-language question with the schema it has already learned from your sources. Its AI maps your words to the right tables, columns and joins, generates the query, runs it, and returns the answer — so you never touch SQL yourself.
When you type a question, iDBQuery doesn't guess blindly. It already holds a memory of your schema — every table, column, type and relationship it found when you connected the source. It uses that, plus a semantic layer that captures what cryptic field names actually mean, to translate your request into precise SQL.
The process runs as a tool-using loop:
- It interprets your question against the known schema.
- It writes and runs the query, checking sample data and statistics when it needs to confirm a column's meaning.
- If a result looks off or incomplete, it refines the query and tries again before answering.
- It can also explain the query it ran, in plain language, so you can see exactly how the number was produced.
Because iDBQuery owns the whole path from question to executed query, you get the analytical power of SQL without writing a line of it. Analysts can read and trust the generated SQL; everyone else can simply read the answer. And every result links back to its source rows, so the SQL isn't a black box you have to take on faith.
Updated 2026-06-22