Demo Result

What thirty seconds bought (D4). Every criterion the grader returned is shown; the overall letter is not, because one answer is not a verdict on a candidate. The radar is the product’s own six-axis profile built from this single answer — five axes are genuinely empty, which is a fact about the data rather than a teaser.

A mixed answer

Uneven scores, which is the normal case. Criterion labels come from resultsCopy — the same ones /results uses, not demo-only wording.

Here's what we picked up

This is based on a single answer

  • STAR structure
    4/5
  • Specificity & detail
    2/5
  • Personal impact
    3/5
Діаграма профілю навичок — текстовий розбір нижче
A full interview fills all six axes
PropTypeDefaultDescription
language'ua' | 'en'—Picks the copy from `demoCopy`; criterion labels come from `resultsCopy`.
criteriaCriteriaScores—Real per-criterion scores from `/api/grade`, 1–5. All of them are shown.
overallGradeLetter—Feeds the radar profile only. Never drawn — one answer is not an overall verdict.
onFull() => void—On to the full interview (the setup form).
onLater() => void—Also the setup form. Nothing is lost either way.

A weak answer

Same tone, same scale, no consolation copy. A demo that flattered would be lying in the one moment the visitor is deciding whether to trust the product.

Here's what we picked up

This is based on a single answer

  • STAR structure
    1/5
  • Specificity & detail
    1/5
  • Personal impact
    1/5
Діаграма профілю навичок — текстовий розбір нижче
A full interview fills all six axes
Sergiy Klymkovych

VoiceMock was built by one designer

I'm Sergiy, a product designer. I built this trainer myself — research, design system, code. On LinkedIn I show how products like this come together, and what it changes in a designer's work.

Product Designer · AI Design Engineer · Designer for Startups

See it on LinkedIn