PrepWise — AI-Powered Interview Preparation Platform

PrepWise helps students and freshers practice mock interviews with AI-generated questions and receive structured, actionable feedback powered by Google Gemini.

Features

• AI Question Generation — Gemini generates role-specific interview questions (HR, Technical Basics, Technical Advanced)
• Timed Interview Sessions — Realistic interview experience with per-question timers (2:45 per question)
• AI Evaluation — Detailed scoring and feedback on each answer with ideal answer suggestions
• Dynamic Feedback — Personalized strengths and improvement areas based on actual performance
• Performance Reports — Downloadable PDF reports with per-question breakdown, scores, and tips
• Analytics Dashboard — Bento grid layout with performance charts, score analysis, category breakdowns, and insights
• Session Persistence — Each interview session stored permanently in Firestore with unique IDs
• Dashboard Stats — Real-time total interviews, average score, best score, and recent sessions
• Authentication — Email/password + Google sign-in via Firebase Auth
• Route Protection — Middleware-based auth guards on all protected pages
• Mobile Responsive — Full mobile support with hamburger navigation

Tech Stack

Layer	Technology
Framework	Next.js 16 (App Router)
Language	TypeScript
Styling	Tailwind CSS 4
Auth	Firebase Authentication
Database	Cloud Firestore
AI	Google Gemini 2.0 Flash
PDF	jsPDF + html2canvas
Icons	Lucide React
Deployment	Vercel (free tier)

Getting Started

Prerequisites

• Node.js 18+
• Firebase project with Auth + Firestore enabled
• Google AI Studio API key (free tier)

Setup

# Clone the repository
git clone https://github.com/Shriya-25/Prepwise.git
cd Prepwise

# Install dependencies
npm install

# Copy environment variables
cp .env.example .env.local
# Fill in your Firebase and Gemini API keys

# Run development server
npm run dev

Open http://localhost:3000 in your browser.

Environment Variables

See .env.example for all required variables. At minimum you need:

• Firebase client config (6 NEXT_PUBLIC_FIREBASE_* vars)
• GEMINI_API_KEY for AI features

Firestore Security Rules

Deploy these rules in Firebase Console → Firestore → Rules:

rules_version = '2';
service cloud.firestore {
  match /databases/{database}/documents {
    match /sessions/{sessionId} {
      allow read: if request.auth != null && resource.data.userId == request.auth.uid;
      allow create: if request.auth != null && request.resource.data.userId == request.auth.uid;
      allow update, delete: if false;
    }
  }
}

Project Structure

├── app/
│   ├── (protected)/       # Auth-required pages
│   │   ├── dashboard/     # Post-login home with stats
│   │   ├── setup/         # Interview configuration
│   │   ├── interview/     # Live interview session
│   │   ├── processing/    # AI evaluation + Firestore save
│   │   ├── feedback/      # Quick score summary
│   │   ├── report/        # Detailed per-question report
│   │   ├── analytics/     # Performance charts & history
│   │   ├── profile/       # User profile management
│   │   └── aptitude/      # Learning resources
│   ├── api/
│   │   ├── generate-questions/  # Gemini question generation
│   │   └── evaluate/            # Gemini answer evaluation
│   ├── components/
│   │   ├── site-navbar.tsx      # Global navigation
│   │   └── site-footer.tsx      # Global footer
│   ├── login/
│   ├── register/
│   └── page.tsx           # Landing page
├── lib/
│   ├── firebase.ts        # Client-side Firebase init
│   ├── firebase-admin.ts  # Server-side Firestore (optional)
│   ├── gemini.ts          # Gemini AI service
│   └── auth-session.ts    # Server-side auth helpers
├── utils/
│   └── interview-setup.ts # Shared role options
├── firestore.rules        # Firestore security rules
├── middleware.ts          # Route protection
└── .env.example

User Flow

Landing Page → Login/Register → Dashboard
  ↓
Setup Interview (choose role + company)
  ↓
AI Generates 10 Questions (HR + Technical)
  ↓
Answer Questions (timed, 2:45 each)
  ↓
Processing (AI evaluates + saves session to Firestore)
  ↓
Feedback Summary (score + badge)
  ↓
Detailed Report (per-question breakdown + PDF download)
  ↓
Dashboard & Analytics (all sessions tracked permanently)

Session Data Model

Each session stored in Firestore contains:

• userId — Owner of the session
• role — Target job role
• company — Target company (optional)
• overallScore — 0-100 score
• scores — Per-category breakdown (HR, Technical Basics, Technical Advanced)
• strengths — AI-generated key strengths
• improvements — AI-generated areas to improve
• summary — Overall performance summary
• recommendedNextSteps — Actionable next steps
• questions — Full Q&A with scores, feedback, and ideal answers
• createdAt — Timestamp
• questionsCount — Number of questions

Deployment

Deploy to Vercel with zero configuration:

npm run build  # Verify build passes
# Then connect your GitHub repo to Vercel

Set all .env.example variables in Vercel's Environment Variables settings.

License

MIT