// feature · interview

Practice for the job you actually saved.

Two ways to practice. A 15-minute live voice call with an AI interviewer who asks questions from your saved JD. Or a 15-question async text round you can run in any quiet moment. Both produce a structured post-session report with score, readiness level, strengths, gaps, and next steps. Shareable as an anonymous public link.

In · Saved job + you
Senior Frontend at Stripe
JD already parsed on file
Master resume on file
Pick async or voice
Click start
Practice
Out · Structured report
Score: 72 of 100
Readiness: ready with prep
Strengths: 4 named
Gaps: 3 named
Next steps: 3 specific
// how it works

How to use AI mock interview.

4 steps. Each one runs in seconds. Scroll in order.

  1. Step 01

    Save the job and pick the mode

    A saved job in your tracker has the JD parsed into structured skills. Open the prep tab and choose async text (15 questions you type) or voice (15-minute live call with the AI interviewer). Async runs in any quiet moment. Voice runs in your browser, no app to install.

  2. Step 02

    Run the round

    Async: 15 behavioral and role-specific questions, you type answers in your own time. Voice: the AI interviewer asks questions live, you answer aloud, the call is bounded by a 15-minute hard cap and a 2-hour daily total across all voice sessions.

  3. Step 03

    Get the structured post-session report

    Score 0 to 100. Readiness level (not ready, needs prep, ready with prep, ready). Strengths and gaps named specifically. Next steps written as concrete actions, not generic advice. The report waits for you on the job page.

  4. Step 04

    Download as PDF or share an anonymous link

    The report downloads as a PDF for your records. Share-link option strips the company name and job title by default so you can send it to a mentor without revealing where you are interviewing. Search engines do not index shared links.

// worked example

A real voice session, abbreviated

One real example, same input reshaped two ways.

// saved job + practice mode
Senior Frontend Engineer at Stripe. Voice mode. 15-minute live call with the AI interviewer. Questions were behavioral plus two role-specific (React performance, system design for a payments form).
// post-session report (abbreviated)
Score: 72 / 100. Readiness: ready with prep.

Strengths:
  - Clear STAR structure on the migration story.
  - Concrete metrics on the performance work (LCP from 4.1 to 2.3).
  - Confident pacing throughout, no rushed answers.
  - Tied team impact to one specific user-facing outcome.

Gaps:
  - Vague on the GraphQL question. You said "I have used it" but no story.
  - Rushed the system-design trade-offs. You named two options but did not weigh them.
  - One filler-word loop in the closing.

Next steps:
  - Prepare one specific GraphQL story (challenge, your approach, outcome).
  - Practice the system-design verbal walkthrough once before the real interview.
  - Pause for 1 second before answering hard questions instead of starting mid-thought.
TakeawayA real interviewer cannot tell you this much without a debrief. The structured report does, on every session.
// under the hood

The mechanics nobody hides.

You should be able to tell how a feature works before you trust it with your job search.

01

Per-job context, not a generic question bank

Voice mode reads your saved JD AND your master resume on file, so the questions are about THIS job, not a generic "tell me about yourself" loop. Async mode reads the JD (skills, experience level, role context) but not the master resume today. Both modes are tied to the saved job in your tracker.

02

Live voice runs in the browser

A real-time AI voice model with bidirectional streaming. No app to install, no Zoom link, no scheduling. Click start, the AI greets you, the call runs for up to 15 minutes. The session is bounded by a real-time audio API constraint at 15 minutes per call and a 2-hour daily cap across all your voice sessions.

03

Separate post-session evaluation pass

After the call ends (or after you submit the async answers), a second AI pass evaluates the full transcript and produces the score, readiness level, strengths, gaps, and next steps. Voice and async use the same evaluation prompt so the report shape is consistent.

04

Anonymous shareable report link

One click generates a public URL like resimay.ai/interview-report/abc123. The company name and job title are stripped from the public view by default, so you can share with a mentor without revealing where you are interviewing. The link is revocable. Search engines do not index it.

05

Audio bytes are not stored

We log byte counts and metadata for the cap-enforcement, not the raw audio. There is no audio playback after the session. The transcript is what survives.

// guardrails

What it deliberately does not do.

Honest limits read as trust signals. Hiding them does the opposite.

  • We do not store your audio. Byte counts only, no playback.
  • We do not share company names or job titles on the public report link by default. You can opt them back in per link.
  • Voice mode is bounded by a 15-minute hard cap per session (a real-time audio API constraint) and a 2-hour daily total across all voice sessions.
  • No live scoring during the call. The evaluation runs post-session only.
  • No video. Audio only on voice mode.
  • English only on both modes.
  • We are not an "undetectable interview copilot." This is practice, not coaching during the real interview.
// questions

Common questions.

If your question isn't here, email support and you'll hear back from the founder.

Async or voice, which should I pick?

Async if you want to think before answering, run multiple rounds, or practice on your phone during a commute. Voice if you want to practice your actual delivery and pacing under live pressure. Most people use both: async earlier in prep, voice closer to the real interview.

How does the voice mode work technically?

A real-time AI voice model with bidirectional streaming, all browser-based. No app, no scheduling. The 15-minute hard cap is a real-time audio API constraint we cannot extend; the 2-hour daily total exists so a stuck session or scripted abuse cannot run unbounded.

What happens to my audio?

We do not store it. We log byte counts to enforce the daily cap, nothing else. There is no playback after the session. The transcript is what survives, and it powers the post-session evaluation.

Can I share my report with someone?

Yes, with one click. The report has a Share button that generates an anonymous public link. The company name and job title get stripped automatically so your mentor sees the feedback but not which job you applied to. The link is revocable any time. Search engines do not index these links.

How much does it cost?

Free accounts: 1 lifetime voice session + 3 async rounds per month (shared with resume tailor, cover letter, and rejection analysis). Pro Weekly, Pro, Pro Quarterly, and Founders include 3 monthly voice sessions + unlimited async. The Voice Pack add-on adds 15 voice sessions per month for $12.99 one-time.

Why does voice mode read my master resume but async does not?

Voice mode passes your master resume to the real-time AI so the interviewer can ask about your actual experience ("tell me about the migration you led at Stripe"). Async mode is shipped without that path today; questions come from the JD context only. Wiring the master resume into the async path is on the roadmap.

// try ai mock interview

Stop reading about it. Run it on a real job.

Create a free account in under a minute. First job tracked, first tailored resume, and first keyword breakdown all happen inside the onboarding flow.

No credit card · 2-minute setup · Cancel anytime