How Our Fraud Screening Actually Works

Direct messaging is the core of Kestrion, which means it's also the first place bad actors would try to show up if we let them. So every first message a recruiter sends to a candidate passes through an automated screening step before it's delivered — not after a report, before it ever reaches the candidate's inbox.

The screening looks at a combination of signals: whether the sending account is a verified recruiter tied to a real company domain, whether the message content matches patterns we've seen in fraudulent job offers (upfront payment requests, equipment purchase demands, off-platform contact pressure), and whether the account's behavior looks consistent with a real hiring process rather than a mass-messaging campaign.

Messages that pass screening are delivered immediately, which is the outcome for the overwhelming majority of first contact on the platform — most recruiters are exactly who they say they are. Messages that trip a signal are held for human review by our Trust & Safety team rather than blocked outright, since automated systems make mistakes in both directions and we'd rather have a person double-check than wrongly block a legitimate recruiter.

This is also why we ask recruiters to keep initial contact inside Kestrion's messaging system rather than moving straight to email or phone — it's the only way the screening step can do its job. Once a conversation is established and both sides have engaged, later messages don't need the same scrutiny.

We treat this system as a living one, not a fixed set of rules. As new patterns of abuse show up anywhere in the industry, we add signals for them. If something ever looks off in a message that did make it through, reporting it — even after the fact — helps us tune the system for everyone else.