IntervBit

An AI‑powered platform for practicing job interviews in real time.

IntervBit helps you sharpen your interview skills through live, AI‑driven voice interviews and post‑interview feedback. Built with Next.js and powered by Firebase and VAPI AI, it simulates realistic interview scenarios and provides actionable insights.

Features

• Real‑Time AI Voice Interviews
  Conduct a live, voice‑based interview with an AI interviewer powered by OpenAI’s GPT‑4, Deepgram for transcription, and 11labs for speech synthesis.
• Automatic Feedback Generation
  After each interview, receive structured feedback across multiple categories (Communication, Technical Knowledge, Problem Solving, Cultural Fit, Confidence).
• User Authentication & Data Persistence
  Sign up or log in with Firebase Authentication; interview transcripts and feedback are stored securely in Firestore.
• Customizable Interview Templates
  Define your own question flows or choose from preset templates; support for various roles, experience levels, and tech stacks.
• Modern UI/UX
  Responsive design using Tailwind CSS animations, Radix UI components, and iconography from Lucide React.

Tech Stack

• Framework: Next.js 15 (App Router)
• Language: TypeScript, React 19
• Styling: Tailwind CSS v4, clsx, class-variance-authority
• Authentication & Database: Firebase (Auth, Firestore)
• AI & Transcription:
  • VAPI AI SDK (@vapi-ai/web) for workflow orchestration
  • OpenAI GPT‑4 for interview logic
  • Deepgram for speech‑to‑text
  • 11labs for text‑to‑speech
• Utilities: react-hook-form, zod, dayjs

Getting Started

Prerequisites

• Node.js v18 or later
• npm (or Yarn)
• A Firebase project with service account credentials
• VAPI AI credentials and a defined workflow ID

Clone & Install

git clone https://github.com/Kumaraditya18/IntervBit.git
cd IntervBit
npm install
# or
yarn install

Environment Variables

Create a .env.local file in the project root with the following:

# Firebase Admin (Server-side)
FIREBASE_PROJECT_ID=your_project_id
FIREBASE_CLIENT_EMAIL=your_client_email
FIREBASE_PRIVATE_KEY="-----BEGIN PRIVATE KEY-----\nYOUR_KEY_CONTENT\n-----END PRIVATE KEY-----\n"

# VAPI AI (Client-side)
NEXT_PUBLIC_VAPI_WORKFLOW_ID=your_vapi_workflow_id

Run Locally

npm run dev
# or
yarn dev

Visit: http://localhost:3000

Usage

1. Sign Up / Log In via the Firebase auth form
2. Choose an interview template
3. Start a live interview by clicking “Call”
4. Speak your responses
5. Receive instant feedback post interview

Deployment

Vercel

• Import repo on Vercel
• Add environment variables
• Deploy with Vercel’s CI/CD

Or self-host with:

npm run build
npm start

To Be Added

• 🔒 Logout functionality for signed-in users
• 📝 Editable feedback and summary export
• 📊 Dashboard with historical performance and metrics
• 🌐 Multi-language support
• 📱 Mobile-first UI enhancements

Contributing

1. Fork this repo
2. Create a branch: git checkout -b feature/my-feature
3. Commit and push
4. Open a pull request

License

MIT License