InterviewPal👾 is a full-stack AI-powered voice assistant webapp that conducts real-time mock interviews, providing instant feedback and score — all without the need for human interaction. It's designed to help users practice and prepare for real-life interviews at their convenience, choosing the type and difficulty of interview they want!
InterviewPal enables users to:
- Engage in AI-led mock interviews anytime.
- Receive real-time grading and feedback based on their answers.
- Customize the type of interview based on roles or topics.
- Improve communication and response skills in a realistic interview setting.
No human interviewer is required — the AI voice assistant handles everything from asking questions to evaluating responses.
This project utilizes the following technologies:
- Frontend: Next.js, React.js, TypeScript
- Voice Assistant: Vapi API
- AI & NLP:
- GPT-4: For conducting interviews
- Gemini AI & Gemini Flash: For generating feedback and performance analysis
- Backend / Realtime Data: Firebase & Firestore
- Validation: Zod
- User initiates the interview through the web interface.
- The Vapi-powered AI voice assistant communicates with the user.
- Interview questions are generated and managed using GPT-4 via API.
- Once the interview concludes:
- Responses are analyzed using Gemini Flash for scoring and feedback.
- Data is stored and fetched in real-time using Firebase.
- Final feedback and improvement suggestions are shown to the user in a detailed report.
- Voice-based mock interview simulation
- Dynamic question generation based on role
- Instant scoring and performance analytics
- Zero human dependency — fully automated
- Realtime data updates and response evaluation
First, run the development server:
npm run dev