Using Teamcast
What is Teamcast
AI-led async behavioral interviews with defensible, coverage-only measurement — for recruiting teams and their ATS/HRIS partners.
Teamcast runs behavioral interviews with Maya, Teamcast's AI interviewer. A recruiter (or a partner integration via the API) submits a role and a candidate. Maya conducts a live, adaptive voice interview, and the platform returns a structured, deterministic assessment. A human approves the plan and the assessment at every step — Teamcast never runs unattended end to end.
How it works
A job defines the role and its competencies. Adding a candidate kicks off plan generation: Maya (with the Plan Engine) builds a résumé-informed interview plan — ranked competencies and the questions that will probe them. A recruiter reviews and approves that plan before anything is sent to a candidate. Once approved, the candidate is invited, and Maya conducts the interview live. When it ends, deterministic scoring produces a coverage-only assessment, which a recruiter reviews and acts on. Every one of those steps is a human checkpoint — see Interview lifecycle for the full state machine.
Two hard rules
Teamcast is built around two constraints that are enforced in code, not just policy — both are EU AI Act alignment decisions:
coverage_summary — sufficient / partial / insufficient per competency — and a human always makes the actual decision.2. No emotion inference. Nothing about a candidate's emotions, affect, or mood is ever inferred or scored. Behavioral integrity checks (e.g. tab switches, fullscreen exits) are logged, never emotional state.
Who uses Teamcast
Recruiting teams use the web app to create jobs, add candidates, review plans and assessments, and monitor live interviews — see Recruiters → Jobs & candidates. ATS/HRIS partners integrate the same lifecycle through the Integration API (A2A) instead of the UI. Candidates experience a short, conversational voice interview — see Candidates → What to expect.