Interview Scorecard & Debrief Consolidator
Give every interviewer a structured scorecard, then auto-consolidate the whole panel into one debrief with averages, disagreement flags, and a hiring-manager-approved decision - so you hire on evidence, not vibes.
A private tool where each interviewer submits a blind scorecard against a role rubric, the panel's input is consolidated with average score, spread, and disagreement flags, and the hiring manager records an approved hire/no-hire decision - all audited and exportable to CSV.
Before you start
- A free Supabase account
- A free Vercel account
- A free Resend account
- A competency rubric (or the willingness to write one) for at least one role
The problem this kills
Interview feedback shows up everywhere except where you need it. A rating in a Slack DM. Two paragraphs buried in an email. A "let's chat" hallway comment that never gets written down. By the time you sit down to make a call, you're stitching together half-remembered impressions and whoever spoke loudest in the debrief wins.
That's how good candidates get passed over, weak ones get waved through, and every hire becomes a gut-feel gamble you can't defend later. Worse, interviewers see each other's takes before they've formed their own - so the panel quietly converges on one person's opinion instead of giving you independent reads.
This tool replaces all of that with one clean flow: a rubric per role, a blind scorecard per interviewer, and a single consolidated debrief that does the math for you.
What you'll build
A private, login-only web app where:
- A recruiter or hiring manager sets a competency rubric for a role (the skills to assess, the rating scale, and the required questions).
- Each interviewer on the panel submits a scorecard - ratings per competency plus written notes - and crucially cannot see anyone else's scores until they've submitted their own.
- The app consolidates the panel into one debrief: average score per competency, the spread (how much interviewers disagreed), and automatic flags for missing scorecards and big disagreements.
- The hiring manager records the final decision (hire / no-hire) with a written rationale - and advancing the candidate or sending any candidate-facing outcome requires explicit approval.
- Everything is audited and exportable to CSV for your ATS or records.
What's inside the Implementation Plan
The plan is a complete, paste-and-go runbook. It opens by interviewing you about your business - your roles, your competencies, your rating scale, how your panels actually work, and your messiest edge cases - so the tool is tailored to how you hire, not a generic template. It reads a short spec back to you and waits for your thumbs-up before building anything.
From there it walks you, prompt by prompt, through standing up the database, the login, the blind scorecard form, the consolidation math, the approval gate, the audit trail, and the CSV export. Every step ends with a ready-to-copy prompt you paste into your AI coding agent.
The governance it includes (this is the point)
- Login so only your hiring team can get in.
- Row-level security so people only ever see candidates and scorecards for their own organization.
- Blind submission so interviewers can't see others' scores before submitting - real independent reads.
- A complete audit trail - who scored what, who decided what, and when.
- A human-in-the-loop approval gate - the app consolidates and recommends, but a person reviews and approves before any candidate is advanced or contacted.
- Duplicate guards - one scorecard per interviewer per candidate, enforced.
Who it's for
Hiring managers and recruiters who are tired of collecting feedback in scattered emails and Slack DMs and can never compare interviewers consistently. If you've ever made a hire you couldn't fully explain afterward, this is for you. No coding experience required - if you can write a job description, you can build this.
You've got this - paste the first prompt and let the agent interview you.