Check language, model, screenshot, voice, shortcuts, and macOS permissions before interviews
Basic Settings
When to use this page
Use this page for first-time setup, new machines, system updates, and the final pre-interview check before a real session.
Defaults are the recommended starting point
Our default setup is already designed as the recommended starting point for most real interview workflows. For most users, one full rehearsal on defaults plus 1-2 small adjustments is more reliable than changing many settings on day one.
This page focuses on configuration, not interview workflow. If you want the fastest path for coding prompts, voice sessions, or follow-up rounds, continue with Core Features.
Complete these 4 checks first
Language and model
Lock your coding language, interview language, and primary model before the real session. Non-coding roles can focus on interview language and model first.
Screenshot and voice
Verify screenshot mode, microphone, system audio, and the Voice toggle. If needed, update shortcuts before interview day instead of during it.
Shortcut conflicts
At minimum, test hide / show app window, generate response, and the voice toggle. These are the three shortcuts you will rely on most.
Permissions and self-check
On macOS, pay extra attention to Accessibility and Screen Recording. If you are worried about sharing behavior, rehearse once in self-check tools or on the real interview platform.
Recommended default starting path
Keep General for non-coding
If the interview is not coding-heavy, keep the app mode / coding format preference on General. You usually do not need to switch between LeetCode and ACM right away.
Lock one language and one model first
Run your rehearsal with one stable interview language and one primary model. The default priority should be stability, not complexity.
Let voice start on the default workflow
Rehearse one full round on the default voice workflow first, then decide whether Smart route, hotwords, silence split, or auto-trigger rhythm really need tuning.
Start the knowledge base from the cloud resume
If you want answers to sound more like your real background, complete the cloud resume first and then sync it into the knowledge base before uploading lots of mixed documents.
Recommended setup order
Open Settings and do one full pass
Do not only tweak one switch. Review language, model, screenshot, voice, shortcuts, and permissions together.
Lock your primary model and output language
Avoid switching models right before or during a real interview. Stability usually matters more than chasing a "stronger" model last minute.
Verify screenshot and audio devices
If you plan to use screenshot and voice in the same round, confirm microphone, system audio, and screenshot mode together.
Open the shortcut helper card
Use Ctrl/Cmd + / to bring up the shortcut guide and rehearse the keys you
will actually use.
Finish with one real self-check run
Use the same hardware, meeting tool, sharing style, and shortcut habits that you plan to use in the actual interview.
How to configure the key settings
Reading tip
Read these four groups in order. Each tab is organized like a quick settings walkthrough so you can scan it fast before interview day.
Coding language
This controls the default output language for coding tasks. If the role is not code-heavy, this can stay secondary.
Interview language
This affects both live transcription and the model's response language. It matters most in behavioral rounds, project walkthroughs, and follow-up-heavy sessions.
Solving model
Pick one primary model and stick with it. Frequent switching can change tone, answer length, and latency in ways that are hard to control live.
Recommended approach
Rehearse 2-3 questions with one stable setup before the interview. Familiarity usually beats experimentation in the final hour.
Screenshot mode
Choose between full-screen and partial capture based on the question type. Partial capture is often easier for long prompt pages and cluttered browser layouts.
Voice mode
Use this for real-time interview support. Start by validating the default voice workflow and transcription stability with your current microphone or headset combination before tuning Smart route, hotwords, or silence thresholds.
Input devices
Validate microphone source and system audio source before the interview starts, not after you realize a question was missed.
Combined rehearsal
If you will listen and capture visual prompts in the same round, rehearse screenshot and voice together instead of testing them separately.
Keep these toolbar entries
Screenshot, Voice, Knowledge Base, and Settings should all be easy to reach before the interview. Try to avoid reconfiguring them live.
Common settings entry points
Knowledge Base, voice settings, advanced settings, and support are the places you will revisit most often during rehearsal and final checks. In most cases, you will have a smoother experience if you finish the basic layer before opening advanced settings.
Interface prep
Confirm window size, theme, transparency, and panel position before the interview so you are not resizing or moving things under pressure.
Continue to workflow
If you want to see how Screenshot, Voice, and Knowledge Base work together, continue with Core Features.
Accessibility
This is the most important permission for desktop interaction and keyboard-driven behavior on macOS. Re-check it after system upgrades or after moving to a new .app.
Screen Recording
If you want to validate sharing, visibility, and privacy-related behavior, review Screen Recording at the same time instead of checking it later only after capture fails.
Reinstall or moved app reminder
If you reinstalled the app or changed its install location, macOS may
treat it as a new application identity. Re-authorize the actual
installed .app.
Desktop behavior checks
If desktop clicks, focus, or window behavior still feels unusual, also inspect related macOS desktop settings instead of assuming the app is the only cause.
Key reminder before the interview
New machine, newly packaged app, or recent macOS upgrade are the three most common reasons permission state changes unexpectedly. Re-check them before the interview starts instead of during it.
60-second pre-interview check
- Open Settings and confirm language, model, screenshot mode, and voice were not changed by accident.
- Press
Cmd/Ctrl + Bto hide or show the app window,Cmd/Ctrl + Enterto submit, andCmd/Ctrl + Mto toggle live voice transcription. - Run one minimal check in self-check tools or a test meeting.
- On macOS, re-check permission state if this is a new machine, a new build, or a recently upgraded system.
Where to look first if something feels off
Shortcuts do not respond
Start by checking whether your high-frequency shortcuts conflict with the OS, input method, or meeting software.
Sharing or display feels wrong
Start with platform compatibility, self-check flow, sharing setup, and window visibility issues.
Want to understand mode combinations
Review how Screenshot, Voice, and Knowledge Base work together during a real interview instead of only tweaking settings.
Want to improve the overall feel
Check window placement, hide / show app window behavior, and platform-specific usage suggestions.
AI Reading Assistant
Send to your preferred AI
Smart Summary
Deep Analysis
Key Topics
Insights