Meet the RareFourth Interviewer
The AI that runs the whole interview — and never takes its eyes off it.
RareFourth's interviewer is an autonomous AI that conducts, watches, and evaluates a candidate's interview from start to finish. It asks the questions, listens to the answers, adapts as it goes, and quietly builds a fair, evidence-based verdict behind the scenes. The candidate simply answers. The hiring team gets the decision.
The idea
What it is
Think of it as a neutral, tireless interviewer that treats every candidate exactly the same. It doesn't get tired, doesn't play favourites, and doesn't have off days. It doesn't guide, hint, or correct — it observes, questions, and judges.
It runs on its own. There's no panel to schedule, no interviewer to prep, and no scorecard to fill in from memory afterwards. One link, and the interview happens.
The flow — four simple stages
How it works
It prepares.
Given the role and the skills that matter, the AI builds a tailored set of questions — a mix of conceptual, practical, and hands-on coding tasks — pitched to the seniority you're hiring for.
It interviews.
The AI asks questions, listens (typed, spoken, or written in code), and talks back in a natural voice. Everything runs in the candidate's browser — no installs.
It adapts.
This is the part a script can't do. When a candidate is doing well, it pushes harder. When they're struggling on a topic, it shifts ground and tries another. It reads the room and follows the candidate — just like a great human interviewer would.
It evaluates.
As answers come in, the AI scores each one against a consistent rubric — correctness, depth, problem-solving, communication, and more — and assembles a private scorecard with strengths, weak points, and a clear hire / no-hire recommendation.
Integrity
What it watches for
Because the interview is unsupervised, the AI keeps the room honest. Using the candidate's camera (and an optional second phone camera for a side angle), plus signals from the browser, it looks out for:
- Someone else in the room — a second person, or help leaning in off-camera.
- A phone or extra device in hand or on the desk.
- A second screen or monitor being used on the side.
- The candidate leaving the frame or looking off to read something.
- Tab switching, exiting full-screen, or pasting large blocks of answer.
- Background voices coaching from off-screen.
Anything suspicious is captured as evidence — a saved frame or a logged event — so nothing is a matter of opinion. Clean moments aren't stored; only what matters is kept.
Important: these signals are indicators, not accusations. They're flagged for a human to review, never used to auto-reject anyone.
The verdict
How it judges — fairly
Every candidate faces the same standard, scored the same way. No first-impression bias, no inside tracks. Each score traces back to something the candidate actually said, wrote, or did — so when the AI recommends a hire, the reasoning is right there to see.
The AI produces a recommendation. The decision stays with your team, who can review the questions, the answers, the scoring, and every piece of evidence before choosing.
Who sees what
The candidate sees
Questions, a timer, and a coding space. Nothing else — no scores, no feedback, no hints.
The hiring team sees
Everything — the full scorecard, the red flags, the weak points, the evidence, and a recommendation.
Why it matters
Fairer interviews. Faster shortlists. Decisions you can actually defend — backed by evidence, not gut feel.
The interview, done right — so you can just decide.