Intervues← Journal

Why a patient interviewer matters

Most AI interviewers treat silence as failure. We built the opposite — a five-tier silence ladder calibrated for Indian English.

Most people don't fail interviews on knowledge. They fail on the translation — knowing it, then saying it out loud, under mild pressure, in real time.

That's why Intervues waits. Not because we're slow, but because a good interviewer waits. "Umm" and "no?" at the end of a sentence aren't stalling — they're thinking. Other tools run a 30–50% word-error rate on Indian-accented English and fire an answer on every "mmm." We run a five-tier silence ladder with about 25% more pause budget.

What the ladder actually does

The point isn't to make the interview easier. It's to make it honest — so what you practise is what you'll face, and what we score is what you actually said.

Why this shows up in the report

Every score in the fifty-page report is cited to a timestamp and a quote. If we cut you off at second two, the report would be scoring a performance you never got to give. Patience isn't a UX nicety. It's the measurement instrument.

We're building in the open. More soon.