Skip to main content

​
Prerequisites

Before you begin, ensure you have:
  • Node.js (LTS version recommended)
  • A MongoDB database (MongoDB Atlas free tier or local instance)
  • A Google Gemini API key (get one here)

​
Setup in 4 steps

1

Clone and install dependencies

Clone the repository and install packages for both frontend and backend:
git clone <repository-url>
cd resume-generator

# Install backend dependencies
cd backend
npm install

# Install frontend dependencies
cd ../frontend
npm install
2

Configure environment variables

Create a .env file in the backend/ directory with your credentials:
MONGO_URI=your_mongodb_connection_string
JWT_SECRET=your_long_random_secret
GOOGLE_GENAI_API_KEY=your_gemini_api_key
Never commit your .env file to version control. If you accidentally expose your API keys, rotate them immediately.
Getting your MongoDB URI:
  • For MongoDB Atlas: Navigate to your cluster → Connect → Connect your application → Copy the connection string
  • For local MongoDB: Use mongodb://localhost:27017/resume-generator
Generating a JWT secret:
node -e "console.log(require('crypto').randomBytes(32).toString('hex'))"
3

Start the development servers

Open two terminal windows and start both servers:Terminal 1 - Backend:
cd backend
npm run dev
The API server will start at http://localhost:3000Terminal 2 - Frontend:
cd frontend
npm run dev
The React app will start at http://localhost:5173
The backend uses nodemon for hot reloading, so changes to server code will automatically restart the server.
4

Create your account

  1. Open http://localhost:5173 in your browser
  2. Click Register to create a new account
  3. Enter your email and password (minimum 6 characters)
  4. You’ll be automatically logged in after registration
The authentication system uses JWT tokens stored in httpOnly cookies for security:
app.use(cors({
  origin: "http://localhost:5173",
  credentials: true
}))
Source: backend/src/app.js:9-12

​
Generate your first interview report

Now that you’re set up, let’s create your first interview report:
1

Prepare your materials

You’ll need:
  • A job description (paste the full text from a job posting)
  • Either a resume file (PDF or DOCX) OR a self-description
For best results, upload a text-based PDF or DOCX resume. Scanned/image-only PDFs may not parse correctly, though the system will attempt to send them directly to Gemini as a fallback.
2

Submit the form

On the home page:
  1. Paste the job description in the left panel (required)
  2. Upload your resume (PDF/DOCX, max 5MB) in the right panel, OR
  3. Write a self-description if you don’t have a resume ready
  4. Click Generate My Interview Strategy
The form validates your input:
if (!jobDescription.trim()) {
  window.alert("Please paste the job description first.")
  return
}
if (!resumeFile && !selfDescription.trim()) {
  window.alert("Please upload a resume (PDF/DOCX) or fill in the self description.")
  return
}
Source: frontend/src/features/interview/pages/Home.jsx:17-24
3

Wait for AI analysis

The system will:
  1. Extract text from your resume (if provided)
  2. Send your profile and job description to Google Gemini
  3. Generate a structured interview report with match score, questions, skill gaps, and preparation plan
This typically takes 20-30 seconds. You’ll see a loading screen:
<main className='loading-screen'>
  <h1>Loading your interview plan...</h1>
</main>
Behind the scenes, the API makes this call:
const response = await ai.models.generateContent({
  model: "gemini-3-flash-preview",
  contents,
  config: {
    responseMimeType: "application/json",
    responseSchema: zodToJsonSchema(interviewReportSchema)
  }
})
Source: backend/src/services/ai.service.js:63-70
4

Review your report

Your interview report includes:
  • Match Score: A 0-100 score indicating how well your profile fits the role
  • Technical Questions: Role-specific questions with interviewer intentions and answer strategies
  • Behavioral Questions: Soft-skill questions with suggested response frameworks
  • Skill Gaps: Areas where your profile doesn’t fully meet requirements, ranked by severity (low/medium/high)
  • Preparation Plan: A day-by-day roadmap with specific tasks to prepare effectively
All reports are stored in MongoDB and accessible from your dashboard:
const interviewReport = await interviewReportModel.create({
  user: req.user.id,
  resume: resumeText,
  selfDescription: resolvedSelfDescription,
  ...interViewReportByAi,
  title: resolvedTitle,
  jobDescription: resolvedJobDescription
})
Source: backend/src/controllers/interview.controller.js:109-117
5

Export your tailored resume (optional)

From the interview report page, you can generate a tailored resume PDF optimized for the specific job:
  1. Click the Generate Resume PDF button
  2. The system will create an ATS-friendly HTML resume using AI
  3. Convert it to PDF using Puppeteer
  4. Download the file
The generated resume:
  • Highlights relevant experience for the target role
  • Uses professional formatting with subtle color accents
  • Is optimized for Applicant Tracking Systems (ATS)
  • Fits within 1-2 pages
API endpoint:
POST /api/interview/resume/pdf/:interviewReportId
Source: backend/src/controllers/interview.controller.js:172-193

​
Common issues

Error: MongoServerError: bad authSolutions:
  • Verify your MONGO_URI in .env includes the correct username and password
  • For MongoDB Atlas, whitelist your IP address:
    1. Go to Network Access in Atlas
    2. Click “Add IP Address”
    3. Either add your current IP or use 0.0.0.0/0 for development (not recommended for production)
  • Ensure your MongoDB cluster is running
Error: secretOrPrivateKey must have a valueSolution:
  • Ensure JWT_SECRET is set in backend/.env
  • Restart the backend server after adding the environment variable
  • The secret should be at least 32 characters long for security
Error: Unable to read the uploaded PDFCauses:
  • The PDF is image-based (scanned document without text layer)
  • The PDF is encrypted or password-protected
  • The file is corrupted
Solutions:
  • Re-export your resume as a text-based PDF from Word or Google Docs
  • Convert the PDF to DOCX and try uploading that instead
  • Use the “Quick Self-Description” field as a fallback
The system attempts a fallback:
// If local parsing fails, send raw file to Gemini
if (aiMimeType) {
  resumeFileForAi = {
    mimeType: aiMimeType,
    data: req.file.buffer.toString("base64")
  }
}
Source: backend/src/controllers/interview.controller.js:70-75
Error: Access to XMLHttpRequest blocked by CORS policySolution:
  • Ensure both servers are running on the correct ports (backend: 3000, frontend: 5173)
  • Check that the frontend is making requests to http://localhost:3000
  • Verify CORS is configured with credentials in backend/src/app.js

​
Next steps

Installation guide

Detailed setup instructions for development and production environments

Authentication API

Complete documentation of all backend endpoints and request/response schemas

Build docs developers (and LLMs) love

Get started for freeTalk to us