Intervue - AI-Powered Interview Practice

Intervue is a full-stack web application that helps users practice interviews using AI-powered question generation and evaluation. The application uses Google's Gemini 1.5 Pro API for generating interview questions and evaluating responses.

Features

• Dynamic interview question generation based on job role and interview type
• Real-time speech-to-text input (Web Speech API)
• Follow-up questions: If your answer is vague, the AI will ask a follow-up and lock your initial answer
• Distinct answer fields: Main and follow-up answers are kept separate and editable only at the right time
• Timer per question: Customizable (minutes and seconds), auto-advances to next question/report when time is up
• Retry logic: Reset main or follow-up answer independently, with timer behavior tailored to each
• Comprehensive evaluation report: See feedback and download as PDF
• Modern UI/UX: Material UI, smooth transitions, bold section headers, toggle buttons, and theme support
• Dark/Light mode toggle
• Responsive design
• Smooth animations and transitions

Screenshots

Landing Page

Interview Page

Report Page

Tech Stack

• Frontend: React, Material UI, Framer Motion
• Backend: Node.js, Express
• AI: Google Gemini 1.5 Pro API
• Speech Recognition: Web Speech API
• PDF Generation: html2pdf.js

Prerequisites

• Node.js (v18 or higher)
• npm or yarn
• Google Gemini API key

Setup

1. Clone the repository:

   git clone https://github.com/06sarv/intervue.git
   cd intervue

2. Install dependencies:

   npm install

3. Create a .env file in the root directory and add your Gemini API key:

   REACT_APP_GEMINI_API_KEY=your_api_key_here
   

4. Start the development server:

   npm start

5. Open http://localhost:3000 to view the application.

Usage

1. On the landing page, enter:

   • Job Role
   • Interview Type (HR or Technical)
   • Number of Questions (1-5)
   • Time per Question (min, sec)

2. Click "Start Interview" to begin.

3. For each question:

   • Type your answer or use the microphone button for speech input
   • If a follow-up is triggered, your main answer is locked and you answer the follow-up in a new field (mic and typing supported)
   • Use the retry button to reset only the relevant answer (main or follow-up)
   • Timer auto-advances to the next question/report when time is up
   • Click "Next Question" when ready

4. After completing all questions, view your evaluation report.

5. Download the report as a PDF or start a new interview.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Made with ❤️ by Sarvagna