R Rubric

Getting Started

Rubric is a desktop application for running 30-minute system design interviews. It bundles a curated library of interview scenarios, a four-panel live-session workspace, weighted rubric scoring, and AI-authored reporting into a single calm tool that an interviewer can keep open beside a video call.

This page is the shortest path from a fresh install to a finished interview report.

What Rubric is for

Rubric is built for the person running the interview, not the person taking it. The candidate sees only a shared whiteboard. You see the whiteboard, a private interviewer guide, a structured rubric, and a notes pane that timestamps every line you write.

Use it for hiring rounds, mock interviews with juniors on your team, or to dry-run a question before you put it in front of a real candidate.

What you will need

  • A Mac, Windows, or Linux desktop. macOS is the most polished target; Windows and Linux are supported.
  • An API key from one of the supported providers if you want AI features:
    • Anthropic (Claude) — recommended.
    • OpenAI (GPT models).

The AI features (interview generation, report writing) need a key. Live sessions, scoring, and the bundled library work fully offline.

Install

  1. Download the installer for your platform from the Download page.
  2. Run the installer.
    • macOS: open the .dmg and drag Rubric into Applications.
    • Windows: run the signed installer.
    • Linux: make the AppImage executable (chmod +x Rubric.AppImage) and double-click.
  3. Launch Rubric. The first window is the Library.

The MIT LICENSE is included in every install package.

First-run setup

Open Settings from the library sidebar. Three things are worth setting up before you run a real session:

  1. Provider — choose Anthropic or OpenAI.
  2. API key — paste it. The key is stored in a local database on your machine and never leaves it except to call the provider directly.
  3. Default storage — pick where generated interviews and reports get saved. Local filesystem is the simplest start. See Storage Backends for trade-offs.

You can change any of this later. The app still runs without an API key — you just can’t generate new interviews or write reports.

Run your first interview

  1. From the Library, pick an interview. The detail screen shows the description, difficulty, and target roles.
  2. Click Start Session. Enter the candidate’s name.
  3. The four-panel Live Session opens. Share the top-left whiteboard with the candidate (screen-share or send the link).
  4. As the candidate works, follow the Interviewer Guide in the top-right: ask the suggested follow-ups, watch for the listed red flags.
  5. Capture observations in the bottom-right Notes panel. Each note is automatically timestamped against the elapsed clock.
  6. Score in real time in the bottom-left Rubric panel — or wait until the end. Either is fine.
  7. When time is up, click End Session. Rubric drafts the report and opens it for review.

That is the loop. Everything else in this guide is detail on top of those seven steps.

Where to go next