The rules on AI in hiring change from one country to the next. Rather than handing you a pile of switches and hoping you set them right, we hold everyone to whichever rule asks the most of us.
Every one of these is on the record per person, with the version of the wording they were shown.
Applicants are told in plain language what the AI looks at and how, and they agree to it before a single resume is read. What they agreed to, and which version of the wording they saw, is kept.
Every outbound call opens by saying it's an AI, that it's being recorded, and asking whether that's alright. Someone who says no is offered a person instead. No custom prompt anyone writes can turn that off.
Say it once and the pending calls are cancelled, the application moves out of the calling queue, and nothing dials that number for you again.
Somebody can turn down one job, or your company, or everything. Whichever they pick is checked at the moment a message would go out.
We collect no demographic data. None.
There is nowhere in this product that race, sex or age is stored, because there is no field for it. That also means we can't produce a demographic bias audit: you can't analyse what was never collected.
Where someone lives never touches their score.
Location filters your search, and that's all it does. It isn't one of the things the match score is calculated from, and postcodes are stripped out of resumes before scoring.
A candidate can ask for everything, or for nothing.
Anyone can download everything we hold on them, or ask us to delete it. A person reviews every deletion, and once it is done we can still show you it happened without having kept any of what we deleted.
A person still signs off every hire.
Nothing here auto-rejects. That's a product decision, not a compliance feature, and it's the reason the reasoning is always on the screen next to the score.
Free to try. No credit card, and no credit spent until you ask for work.