Questions, answered

An AI interview agent listens to your interview in real time, transcribes what's being said, and suggests how to answer — like having a well-prepared friend beside you. Rep-AI does this in a desktop overlay that only you can see.

You can add your CVs, projects, and the vacancy you're interviewing for in the dashboard. Rep-AI includes that context in every prompt, so suggestions reference your actual experience instead of generic advice.

No. The free tier runs entirely on your machine with your own API keys — download the app, paste your keys, and start. An account is only needed if you want managed keys, cloud session sync, or to store CVs and vacancies for context.

Rep-AI uses a premium real-time speech-to-text engine. Provisional text appears as words are spoken and commits as each utterance completes, so you always see what was just said. Multiple languages are supported with automatic detection.

No. The overlay is excluded from screen capture, so it doesn't appear when you share your screen in a video-call app. It's a native desktop window visible only on your machine.

That depends on where you and the interviewer are. Rep-AI processes the audio of a conversation, and many places — including several US states and most of the EU — require everyone on a call to consent before it's recorded. You are responsible for knowing and following the rules that apply to you, and for any employer or platform policy you've agreed to. Rep-AI gives you the tool; the decision to use it is yours.

On the free tier, audio goes to the speech-to-text provider you configured and your prompts go to the AI provider you configured — using your own keys, under your own accounts. Nothing is sent to Rep-AI's servers. On Pro, those same requests are routed through keys we provision for you. Uploading a session for later review is always opt-in.

Not unless you ask for it. Transcripts stay on your machine by default and you can export them at any time. Session upload to your dashboard is opt-in, and you can delete uploaded sessions whenever you like.

Rep-AI runs natively on macOS 13 or later and Windows 10/11 (64-bit). It captures the interviewer through system audio and you through the microphone, and works alongside any video-call app running on your computer.

Yes. Rep-AI captures system audio rather than hooking into a specific app, so it works with any video-call software — including anything running in a browser tab. There's no extension to install.

Rep-AI works with OpenRouter, Groq, and the Claude, Codex and Gemini command-line tools if you already have them installed. You choose the provider and model in settings.

Yes. You can change microphone or output device during a session without dropping the transcription stream — useful if a headset disconnects mid-call.

The desktop app is free to download and use with your own AI and speech-to-text keys, with no account and no minute limits — you pay your providers directly for what you use. Pro adds keys we manage for you, so there's nothing to configure.

Pro is $10 per month or $100 per year and includes 5 hours of interview time per month on managed AI and speech-to-text keys, plus cloud session sync. If you need more than 5 hours, extra time is $3 per hour on demand.

Only time in an active session counts, and only on the Pro tier — sessions on your own keys never consume minutes. Time is measured per session and rounded up to the nearest 15 minutes.

Cancel any time from the billing page in your dashboard. Your plan stays active until the end of the period you've already paid for, and you keep access to the free tier afterwards.