Made with ❤️ at Sundai hacks at MIT.
An interactive AI-powered mock interview tool that lets users speak their answers, receive real-time transcriptions and feedback, and converse with an AI interviewer using OpenAI’s Realtime API.
Built for practicing behavioral interviews with real-time coaching, voice interaction, and intelligent feedback.
- 🎤 Real-time speech-to-text using OpenAI's Realtime Transcription API
- 🧠 GPT-4o interviewer that asks questions, listens to your answers, and provides personalized feedback
- 📊 Smart analysis of your response structure (STAR), clarity, and impact
- 🔈 Text-to-speech (TTS) voice output from the AI interviewer
- 🖥️ Web interface with microphone support and clean UI
- Frontend: Next.js (React) + Tailwind CSS
- Authentication: Clerk
- Realtime Audio: OpenAI Realtime API
Clone the project and create an env file with your OpenAI API key included.