One standard, so every read compares.

Set what each stage should establish. The agent turns that goal and the role's FitCheck traits into assessments with a shared Below, Meets and Exceeds bar.

See how it works

How it works.

Goal in, assessments out

Write one sentence about what the interview should establish. Superposition generates the assessments from that goal, the stage format and the role's traits. Edit any of them.

One goal, in a sentence
Say what the stage should establish. That is the whole setup.

Generated assessments
Each one spells out Below, Meets and Exceeds so every interviewer holds the same bar.

Each interviewer, their own card

Strong yes to Strong no, ratings per assessment, evidence notes. Every scorecard is attributed to the person who wrote it and never overwrites a teammate's.

Your card, your name
Recommendation, ratings and evidence, attributed to you. Nobody overwrites it.

Evidence, not just a rating
Open Evidence on an assessment and write down what the candidate actually said.

See where the team disagrees

An Interviewers disagree label marks the conversations worth reviewing together. Jump to the transcript or recording to settle it on evidence.

Settle it on the transcript
Jump to the timestamp and decide on what was said, not who spoke loudest.

Saving a card moves nobody

A scorecard records a judgment. The next-stage decision is a separate call the hiring team makes after reading everyone's feedback.

Setup guide in the docs

Discover more features

Voice intake Deep web sourcing FitCheck Bring your own Calibration Personalized outreach Replies handled Review queue Grace period Escalations Agent scheduling Signal Interview guides Insights Slack MCP server ATS submission Calendar and mailbox API and security

Get started

Find your perfect weirdo

The person who would fail anywhere else and belongs with you. Tell us the role and we will show you who the agent would go after first.

Start a search