Document Search

What it does

Document Search keeps an account's own documents and finds them again. A document is saved as text with a title; the platform indexes it for full-text search and answers a query with the matching documents ranked by relevance, each with a snippet that shows where the match is and a link back to the whole document.

What exists now

Data handling

Documents are stored in the platform's PostgreSQL database for the account that saved them. Nothing is sent to an outside service: indexing and search run inside the database. Do not save special-category personal data unless you have a lawful basis; the privacy notice states what the platform collects and keeps.

Operating needs

• PostgreSQL; the index is a database index and needs no separate search service.

Limits

• It is keyword search with stemming. It does not understand meaning, summarise, extract fields or answer questions, and it does not use a language model.
• It takes text, not files: it does not read PDFs or images.
• Documents cannot be deleted, so a document saved in error stays; erasure of an account's data follows the data request process.
• Search is per account; there is no shared or organisation-wide index.

Availability

Document Search is running on RJV's own platform. RJV does not sell a licence for it as a separate product today: the commercial terms state what is on sale. It can be part of the result of an engineering engagement, where the written proposal names it, its licence terms and what you would hold at the end. To discuss it, use Work with Us.

Evidence

Indexing, ranking, snippets, account isolation and the refusal to edit or delete are covered by the platform's own tests. The evidence register lists the graded claims.