Frequently asked questions
Everything you need to know about Dusky.
How it works
Will interviewers see Dusky during screen share?
Dusky registers itself as a "screen share excluded" window at the OS level, so it stays out of screen shares and recordings on Zoom, Meet, and Teams. It is one OS-level flag and capture methods differ, so test it on your own setup before a real interview.
Which platforms does Dusky work with?
We test Dusky against Zoom, Google Meet, and Microsoft Teams. It runs as a separate desktop app alongside your meeting rather than as a plugin, so it generally works with other platforms too — try yours before you rely on it.
What is Think Deeper mode?
Think Deeper mode uses extended AI reasoning for complex questions like system design, architecture trade-offs, and multi-step problems. It takes longer than a normal answer. Activate it with Cmd/Ctrl + Shift + Enter when you need a more thorough response.
Which languages can Dusky transcribe?
Six: English, Spanish, German, French, Portuguese, and Italian. The language is detected automatically, turn by turn, so there is nothing to set — and no picker, because on our speech provider's streaming API the language is a property of the model rather than something you choose per session.
Can Dusky answer in a language it can't transcribe?
Yes. The answer language is a separate setting with 22 options, and only the six above can also be transcribed. If you pick one of the others, the app tells you so at the moment you choose it.
How fast is the transcription?
Transcription appears in near real-time. We use streaming speech-to-text that processes audio as it comes in, so words land while the other person is still talking.
Does Dusky work with phone interviews?
Dusky captures audio from your system. For phone interviews, you can use speakerphone or route your phone audio through your computer. For best results, use video calls where audio is already on your computer.
Privacy & data
Where is my data stored?
Audio is streamed to our transcription service and discarded immediately after processing. Transcripts and AI responses are stored only on your device. We never store your interview content on our servers.
What permissions does Dusky need?
Microphone, to transcribe your own voice. Screen recording, to capture the interviewer's audio coming out of your speakers and to take screenshots when you ask for one. Both are standard macOS/Windows permissions you control, and you can revoke them anytime in System Settings.
Can I delete my local history?
Yes. Session data is stored encrypted on your device only. Signing out wipes it — your CV, saved conversations, session history, and screenshots all go. You can also delete the Dusky data folder yourself at any time.
Do you share my data with third parties?
We use AssemblyAI for real-time transcription and leading AI providers (such as Anthropic) for suggestions. Audio and text are processed in real-time and not retained by these providers for training. See our Privacy Policy for full details.
Is my data encrypted?
Yes. All data in transit is encrypted (TLS). Your session history is encrypted at rest on your device — always, with no setting to turn it off.
Pricing & billing
How does the free trial work?
Download and use Dusky for 15 minutes free. No credit card required. All features are available during the trial so you can test everything before purchasing.
What's the difference between passes and subscriptions?
Passes (Weekly, Job Hunt) are one-time purchases that expire after their duration. The Pro subscription renews monthly until you cancel. Choose passes for temporary interview bursts, Pro for ongoing use.
Can I cancel my subscription?
Pro Monthly subscribers can cancel anytime from their account settings. Your access continues until the end of your current billing period. No questions asked.
Do you offer refunds?
All sales are final. We don't offer refunds, which is why we provide a free trial so you can test Dusky before purchasing.
What payment methods do you accept?
We accept all major credit cards through Stripe. We don't currently support PayPal or cryptocurrency.
Technical
What are the system requirements?
macOS 12+ (Monterey or later) on Apple Silicon or Intel. The installer will run on macOS 11, but we test on 12 and later. Windows 10/11 (64-bit) with 4GB RAM minimum. A stable internet connection is required for transcription.
Is the macOS app signed and notarized?
Yes. The macOS app is code-signed and notarized by Apple.
Does Dusky work offline?
No. Dusky requires an internet connection for transcription and AI features. Local history is viewable offline, but live features need connectivity.
Why does the app need screen recording permission?
Two things need it. It is how Dusky captures the interviewer's voice — their audio comes out of your speakers as system audio, not through your microphone — and it is how Dusky takes a screenshot when you ask it to solve what's on screen. Staying out of screen shares is separate and needs no permission at all: that is a window flag Dusky sets on itself.
Support
How do I get support?
Email us at [email protected]. We read every message and reply as quickly as we can.
Dusky isn't staying invisible during screen share. Help!
The exclusion flag is set by Dusky, but the meeting app decides whether to respect it. On Zoom, open Settings > Share Screen > Advanced and set Screen capture mode to "Advanced capture with window filtering". Restart both apps, then run the test again. Older meeting-app versions may need updating.
The transcription quality is poor. What can I do?
Ensure your microphone is set correctly in the app. Use a headset or external microphone for best results. Background noise and multiple speakers can affect accuracy.
Still have questions?
Contact support →