PrepWise

PrepWise is an AI-powered interview preparation platform that helps candidates:

• Generate personalized interview reports from job description + resume + self-description
• Review technical and behavioral interview questions
• Identify skill gaps with severity levels
• Follow a 7-day preparation roadmap
• Generate an updated resume in PDF format tailored to the target job

---

Features

1) Authentication

• User registration and login (JWT + HTTP-only cookie auth)
• Protected routes in frontend
• Logout with token blacklist support on backend

2) Interview Report Generation

• Upload resume PDF + add job description + self description
• AI generates:
  • Match score
  • 5 technical questions
  • 5 behavioral questions
  • Skill gaps (low, medium, high)
  • 7-day preparation plan
• Report is persisted in MongoDB

3) Previous Report History

• Users can fetch all previous interview reports
• Open any specific report by ID
• Ownership-restricted report access

4) Updated Resume PDF Generation

• From an existing interview report, generate an updated role-tailored resume
• AI creates resume HTML
• Puppeteer converts HTML to downloadable PDF

5) Modern Frontend UX

• Public landing page (/)
• Auth pages (login/register)
• Protected workspace for report generation
• Interview report view with sections + history sidebar
• Navbar with user info and logout

---

Tech Stack

Frontend (client)

• React
• React Router
• SCSS
• Axios
• Vite

Backend (server)

• Node.js + Express
• MongoDB + Mongoose
• JWT authentication
• Cookie parser + CORS
• Multer (file upload)
• pdf-parse (extract text from uploaded resume PDF)
• @google/genai (Gemini model integration)
• zod (schema validation)
• puppeteer (HTML to PDF for updated resume)

---

Project Structure

PrepWise/
  client/   # React frontend
  server/   # Express backend

Prerequisites

Make sure you have installed:
- Node.js (v18+ recommended)
- npm
- MongoDB Atlas (or local MongoDB)
- A valid Google GenAI API key

---

## Environment Variables

Create a `.env` file inside `server/`:

```env
PORT=3000
MONGO_URI=your_mongodb_connection_string
JWT_SECRET=your_jwt_secret
GOOGLE_GENAI_API_KEY=your_google_genai_api_key

Do not commit .env files to version control.

---

Install Dependencies

1) Backend

cd server
npm install

2) Frontend

cd client
npm install

---

Run Locally

Start backend

cd server
npm run dev

Backend runs on: http://localhost:3000

Start frontend

cd client
npm run dev

Frontend runs on: http://localhost:5173

---

Main Routes

Frontend

• / → Public landing page
• /login → Login page
• /register → Register page
• /workspace → Protected report generation page
• /workspace/interview-report → Latest/current report view
• /workspace/interview-report/:interviewID → Specific report by ID

Backend API

Auth

• POST /api/auth/register
• POST /api/auth/login
• POST /api/auth/logout
• GET /api/auth/getMe

Interview

• POST /api/interview
  Uploads resume + JD + self description and generates report
• GET /api/interview
  Fetch all reports for logged-in user
• GET /api/interview/report/:interviewID
  Fetch one report by ID (owner only)
• POST /api/interview/resume/pdf/:interviewReportID
  Generate + download updated resume PDF from report context

---

How Resume Upload Works

1. Frontend sends multipart form data:

   • title
   • jobDescription
   • selfDescription
   • resume (PDF file)

2. Backend:

   • extracts PDF text using pdf-parse
   • calls Gemini to generate interview report JSON
   • validates response with Zod
   • stores report in MongoDB

---

How Updated Resume PDF Works

1. Frontend calls:
   • POST /api/interview/resume/pdf/:interviewReportID
2. Backend:
   • loads report for the current authenticated user
   • prompts Gemini to generate resume HTML
   • validates JSON response
   • renders HTML via Puppeteer
   • returns downloadable PDF (application/pdf)

---

Common Issues & Fixes

1) CORS error with credentials

Ensure backend CORS is configured with:

• origin: "http://localhost:5173"
• credentials: true

2) JWT secret error

If you see secretOrPrivateKey must have a value, verify JWT_SECRET is set in server/.env.

3) Resume PDF generation fails

• Ensure GOOGLE_GENAI_API_KEY is valid
• Ensure Puppeteer installed correctly (npm install puppeteer)
• Retry after server restart

4) Login/logout issues

• Backend logout route is POST /api/auth/logout
• Frontend must call it as POST (not GET)
• withCredentials: true should be enabled in Axios instance

---

Scripts

Backend (server/package.json)

• npm run dev → start server in watch mode

Frontend (Vite defaults)

• npm run dev → start dev server
• npm run build → build production
• npm run preview → preview production build

---

Security Notes

• Use strong JWT_SECRET
• Keep API keys private
• Use httpOnly cookies for auth token
• In production:
  • set secure cookie options (secure: true)
  • configure proper CORS origin(s)
  • run behind HTTPS

---

Future Enhancements

• Toast notifications instead of window.alert
• Report filtering/search
• Saved generated resume versions
• Export interview report as PDF
• Analytics dashboard for improvement tracking

---

Author

Dipit Madan
Made with ❤️ by Dipit Madan