The Interview Agent runs five kinds of round — AI voice screening, live coding with real execution, system design on a whiteboard, MCQ and theory, and take-homes — with seniority-aware questions generated from your job post and grounded in your own documents. It proctors throughout: camera, gaze, multi-face and lip-sync checks plus screen-share content detection, with a fifteen-second grace window and a screenshot on every violation.
A conversational voice screen scored across five weighted phases, live coding that actually executes, system design read off the whiteboard by a vision model, MCQ and theory tests, and take-homes graded against your rubric. You choose which apply to the role; the depth follows the seniority.
It builds the question set from your job post and the knowledge base you upload, so it asks about your stack and your problems — not a generic bank. Difficulty adapts live to how the candidate is answering.
Camera, gaze, multi-face and lip-sync checks run alongside screen-share content detection for AI tools, editors and search. A fifteen-second grace window keeps false positives out, and every violation stores a screenshot. Each score links back to the moment in the recording that earned it.
Configured once, grounded in your own documents.
Async and around the clock — no calendars.
Feeding the Screening Agent, recording attached.