What to change before the next interview.

Upload the recording from the one you just sat. You get notes on the story, the specifics, and how you said it. Not your accent.

A voice recorder with a waveform, and an arrow to a page of notes marked in red

How it works

  1. A voice-memo waveform

    01

    Bring the recording you already have

    Voice Memos, Zoom, Meet, or a call recording. Audio files and video with an audio track both work. You upload the file. The app does not record the interview.

  2. A blank name tag

    02

    Name the role, if you want

    A line like “Senior backend engineer” changes the bar. Feedback then judges specificity and depth against that level, instead of grading the answer in a vacuum.

  3. A page of notes with two red marks

    03

    Read one structured write-up

    The report covers how the answer was built, how specific it was, and how it was delivered. It does not label speakers, and it does not comment on your accent.

What the report covers

One pass over the recording, written back as notes you can rehearse against. A two-person file still works. Questions and answers are split by what was said, not by who was speaking.

STAR
Whether the situation, action, and result are actually there, or the story wanders before it lands.
Specificity
Numbers, named tools, and outcomes. Calibrated when you set a role and level.
Ownership
Where “I” and “we” show what you did, and where the credit goes vague.
Filler
The words that pile up when an explanation gets thin.
Pace and hesitation
Speed, pauses before you start, and the gaps in the middle of a sentence.
Tone
Confidence, energy, and whether the delivery stays flat or moves.

Sample report

Tell me about a time you improved a slow system.

Role: Senior backend engineer

STAR. The situation is clear. The result shows up late, after a long tour of the architecture.

Specificity. You name Postgres and a p95 drop from 1.8s to 240ms. That is the right altitude for this level.

Ownership. “We rolled it out” hides what you shipped. Say which part was yours.

Pace. Steady, then a four-second pause right before the number.

Filler. “Kind of” and “basically” cluster when you explain the tradeoff.

Tone. Confident on the result. Flatter when you describe how the team reacted.

Same review, three places

This site is the home for the product. The analyzer is not open here yet. The iPhone and Android apps are on the way, and they will take the same upload.

  • Web

    This page. Upload and the written report will live here.

  • iOS

    Import a recording from the files you already have.

  • Android

    The same flow, on the phone you used to record.