Claims, and the numbers behind them.

• — Document understanding runs on our own infrastructure Your CV is read and indexed on hardware we own. Scoring uses an external model, which this page will say plainly if it changes.
• — Personal details are removed before analysis National ID, e-mail, phone, bank account and date of birth are stripped before any text reaches a model. Employment dates stay, because they carry the experience.
• — Every AI call is measured Each match reports how long it took, how many tokens it used and where it ran. You can see it under your own result.

This page states what is true of the running system, not what we intend. Each claim above is checked against live configuration every time the page loads.