Quickstart
From a fresh install to a live session. Budget about five minutes the first time, and under a minute for every interview after that.
Three-panel flow: interview setup form, the edge hub with Start interview highlighted, and the interview overlay open beside a meeting window.
public/docs/quickstart/flow.pngFive steps
- 1
Sign in and check permissions
Open the app and sign in. Visit Settings → Permissions and confirm microphone and screen capture are green before you rely on them in a real interview. - 2
Build an interview setup
Pick the role, attach the job description and your resume, choose which surfaces you want, and set a model for each. Full walkthrough → - 3
Start from the hub
The hub is the narrow pill docked to the right edge of your screen. Click Start interview. The overlay opens beside your meeting window and transcription begins. - 4
Work the surfaces
Use ⌘1 for behavioral, ⌘2 when a coding pad appears, and ⌘3 when the conversation turns to architecture. Snip anything on screen with ⌘⌥G. - 5
End the session
Click End interview on the hub. The session is saved to your history so you can review it or replay the setup for the next round.
Do a dry run first
Before an interview that matters, spend two minutes on a rehearsal. Start a meeting with yourself in whichever app the interviewer will use, share your screen, and check three things:
- The overlay is absent from the shared view.
- Transcription is picking up audio from the other side of the call.
- A test snip with ⌘⌥G reaches the assistant and comes back.
Side-by-side comparison: the local screen with the overlay visible, and the shared Zoom stream where the overlay is absent.
public/docs/quickstart/dry-run-split.pngSave the setup
Once a setup works, save it. For a multi-round loop at the same company you can reuse it for every stage instead of rebuilding it each time.