Smart FRCS AI An AI examiner for the FRCS oral viva — spoken, examiner-led and scored on the real band scale

Overview

What it is

A surgeon can know the right answer and still lose the viva: it comes out unstructured, too slowly, or falls apart under the second question. Rehearsing aloud usually depends on a consultant with time to play examiner — the scarcest resource a trainee has.

In Smart FRCS AI every clinical case is a station. The candidate chooses an examiner — Alden or Mira, two original examiner identities holding one examination standard — and an answer mode, push-to-talk or continuous. The examiner asks, listens while the answer is transcribed live, and follows up on the candidate's own words. The station ends with a written report on the 4–8 band scale and an annotated transcript to replay. Keyword Practice offers unmetered revision between full stations.

The engineering is in what the candidate never notices: real-time voice through a server-side proxy so no provider key reaches a phone, examiner turns generated within the station's clinical scope, language-model calls that fail over between providers so a station is never lost, and one minutes balance shared across Apple, Google and web purchases. One codebase serves the web app and both mobile apps.

If your product needs to listen, reason and assess — interviews, language, professional training — this is the kind of system I design and build, end to end.

What it does

• Clinical cases delivered as timed stations with a brief and specialty
• Real-time speech-to-text with push-to-talk and continuous answer modes
• Examiner-led follow-up questions generated from the candidate's own answer
• Assessment on the Intercollegiate 4–8 band scale
• Written report with an annotated transcript replay
• Two original examiner identities, Alden and Mira, holding one standard
• Keyword Practice mode for unmetered revision, scored on coverage
• One account across web, iOS and Android, with Apple, Google and web purchases