1. Set the scene
Pick a job family, type your target role, add the organization and paste the job description. The interviewer uses every detail to shape its questions, so a real JD gives the sharpest simulation.
How MockPilot works
No accounts, no API keys, nothing to install. The AI runs server-side, so no credential is ever exposed to the browser.
2. Choose the round
HR screening, Technical, Behavioural or Case assessment. Each round changes the questioning style, depth and follow-up strategy, and you can set how many questions you want.
3. Answer by voice or text
Tap the mic to answer out loud — your speech is transcribed in the browser and sent as your answer. The interviewer speaks its questions back to you; mute the speaker icon any time to go text-only.
4. Turn on camera (optional)
The camera preview helps you rehearse eye contact and posture. Video stays local in your browser and is never uploaded or recorded.
5. Read the coaching signals
After each answer you get clarity, structure and relevance indicators plus one actionable tip. End the session for a full debrief with strengths, gaps, next steps and a readiness score.
Browser requirements
- Voice answers use the browser speech recognition API — best in Chrome, Edge and Safari.
- Allow microphone access when prompted; deny it and text answers still work fully.
- Camera access is optional and purely for self-review.
Design language
Deep teal-black canvas by default, with a light blue theme available from the sun/moon toggle. Frosted glass surfaces, soft ambient glows and compact functional type; motion is limited to breathing states like the recording pulse so the interview stays the focus. Everything is driven by semantic tokens, so the palette can be re-themed in one file.