Every score has a why.
Here is where to find it.

I am an AI interviewer. I conduct live, adaptive voice interviews, score the evidence in each answer against your rubric, and hand your team a ranked shortlist. I am not a decision-maker: I never reject a candidate, and every hiring outcome belongs to a human on your team.

The same interview criteria are applied to every candidate, every time. There are no good-mood interviews and no Friday-afternoon interviews: 100% consistent scoring is the design constraint the rest of the system is built around.

Every score links back to the exact moment in the transcript that earned it. Click a number and you see the evidence — a specific answer, at a specific timestamp. There are no black-box verdicts anywhere in the product.

I don't read down a list. I listen to each answer and choose the follow-up a good interviewer would ask next, depending on what that answer left open — whether anyone else was told, what the candidate would have done without help, what the decision actually cost. Every follow-up I choose, and the reason I chose it, is part of the interview record.

I score against EEO-aware rubrics that read evidence — not accents, names, or where someone went to school. One rubric is used across all 40+ languages and audited for language drift, so nobody scores lower for interviewing in the language they think in.

Scoring drift across demographic groups is measured and reported continuously. Drift gets flagged rather than absorbed.

I never reject anyone — that was never my call to make. Uncertainty, complaints, accommodation requests, and anything the rubric doesn't cleanly cover escalate to your team with the full transcript and my reasoning attached. A borderline candidate becomes a human decision, not a silent no.

Integrity flags work the same way: I detect deepfakes, cloned voices, and AI-generated scripts in real time, but a flag is never an auto-rejection. The interview goes to your team with the evidence, and a human makes the call.

Candidate rights under GDPR and CCPA are honored by default: access, deletion, and portability. Regional data-residency options and deletion SLAs are available, and your candidates are never used to train models.

A human reads every candidate escalation. Anyone who feels something was off in their interview can write to support@ourlanterns.com and we will look at it ourselves.

Hiring is high-risk AI under the EU AI Act, and I'm classified, documented, and logged as high-risk rather than argued around it. Underneath sit SOC 2 Type II, ISO 27001, ISO 27018, ISO 27701, ISO 42001, and the NIST AI RMF.

Every question, answer, and score is exportable — a full audit trail for regulators, works councils, and internal review.