Practice realistic mock interviews with AI using OpenAI, Anthropic, Gemini, Groq, or local Ollama models. Voice conversations, instant feedback, customizable interview flows, and complete privacy with offline support. Open source and developer-friendly.
No reviews yetBe the first to leave a review for Interview Coach
Maker
📌
I'm excited to launch Interview Coach, an open-source AI interview simulator that creates realistic, role-specific mock interviews using the AI models you already trust.
I wanted an interview prep tool that wasn't tied to a single provider, worked with local LLMs, and respected user privacy.
✅ Works with OpenAI, Anthropic, Gemini, Groq, and local Ollama models
✅ Fully offline interview practice with your own LLMs
✅ Natural voice conversations (STT + TTS) on the web dashboard
✅ Realistic role-specific mock interviews
✅ Instant AI feedback after every interview
✅ Privacy-first—your conversations stay on your machine when using local models
✅ Cross-platform CLI and web dashboard
Whether you're preparing for software engineering, product, design, management, or any custom role, Interview Coach lets you practice as many interviews as you want without worrying about API lock-in.
I'd love to hear:
Which LLM would you use?
What interview features are you missing in current tools?
Any feedback on the CLI or web dashboard?
Thanks for checking it out! Happy to answer any questions.
Report
Tried Interview Coach with a local Ollama model and was surprised how natural the voice conversation felt. The instant feedback caught a few filler words I didn't notice myself.
Tested it with a few system design questions and the feedback was actually sharper than I expected. Loving that I can swap between models to compare responses.
Tried Interview Coach with a local Ollama model and was surprised how natural the voice conversation felt. The instant feedback caught a few filler words I didn't notice myself.
@ouzhanwebe6kif Glad you liked it!
Tested it with a few system design questions and the feedback was actually sharper than I expected. Loving that I can swap between models to compare responses.
@dogan6960 Glad you found it useful. Thanks Mate!