How does iDBQuery find the right passage across many documents?
iDBQuery indexes your documents — including scanned PDFs via OCR — and uses semantic (embedding-based) search to find the passages that answer your question, not just keyword matches. It returns the answer with a citation to the specific document and location, so you can open the source and confirm it.
When your "data" is a folder of contracts, reports, invoices, or research papers, finding the right sentence is the whole challenge. iDBQuery treats documents as a searchable, citable source.
- Ingest and OCR. Files are ingested and indexed; scanned or image PDFs are run through OCR so their text becomes searchable rather than locked in a picture.
- Semantic search. Rather than only matching exact words, iDBQuery uses embedding-based retrieval to find passages by meaning — a question about "termination notice period" surfaces the relevant clause even if it's worded differently.
- Cited back to the document. The answer points to the specific document and passage it drew from, so you can open the source and verify, exactly as figures from a database are cited to rows.
This is what lets iDBQuery answer questions over unstructured content and blend it with structured data in one model — asking about a number in a database and a clause in a PDF in the same breath. It's especially useful for teams sitting on large document sets, from legal and finance to research, where the answer exists but is buried across hundreds of files.
Updated 2026-06-22