Interview AI

AI-powered interview preparation app built with React, Vite, Express, MongoDB, and LLM integrations. Users can sign up, upload a resume or add a self-description, paste a job description, and get a personalized interview report with:

• Match score
• Technical interview questions
• Behavioral interview questions
• Skill gap analysis
• Day-wise preparation plan
• AI-generated resume PDF tailored to the target role

The frontend is intended to be deployed on Netlify and the backend on Render.

Tech Stack

Frontend

• React 19
• React Router
• Axios
• Sass
• Lottie animations
• Vite

Backend

• Node.js
• Express
• MongoDB with Mongoose
• JWT cookie authentication
• Multer for file uploads
• pdf-parse for resume text extraction

AI / External Services

• Groq for primary structured generation
• Google Gemini as fallback
• html2pdf.app for resume PDF generation

Project Structure

GenAI/
+-- backend/
|   +-- server.js
|   +-- package.json
|   `-- src/
|       +-- app.js
|       +-- config/
|       +-- controllers/
|       +-- middlewares/
|       +-- models/
|       +-- routes/
|       `-- services/
`-- frontend/
    +-- package.json
    +-- vite.config.js
    +-- public/
    `-- src/
        +-- features/
        +-- app.routes.jsx
        +-- App.jsx
        `-- main.jsx

Features

Authentication

• Register and login with email/password
• JWT stored in an HTTP-only cookie
• Protected routes on the frontend
• Logout support with token blacklisting

Interview Report Workflow

1. User submits a job description.
2. User uploads a resume PDF or writes a self-description.
3. Backend extracts resume text and sends the profile + JD to AI.
4. AI returns a structured interview report.
5. Report is stored in MongoDB and shown in the dashboard.

Resume PDF Generation

• A saved interview report can be converted into an AI-tailored resume PDF.
• The backend generates structured HTML first, then converts it to PDF using html2pdf.app.

Main User Flow

Home page

• Paste job description
• Upload resume
• Optionally add self-description
• Generate a new interview plan
• View previously generated reports

Interview report page

• See match score
• Review technical questions
• Review behavioral questions
• Review skill gaps
• Follow the preparation roadmap
• Download generated resume as PDF

API Routes

Auth routes

• POST /api/auth/register
• POST /api/auth/login
• GET /api/auth/logout
• GET /api/auth/get-me

Interview routes

• POST /api/interview/
• GET /api/interview/
• GET /api/interview/report/:interviewId
• POST /api/interview/resume/pdf/:interviewReportId

Environment Variables

Backend .env

Create backend/.env with:

MONGO_URI=your_mongodb_connection_string
JWT_SECRET=your_jwt_secret
GOOGLE_GENAI_API_KEY=your_google_genai_api_key
GROQ_API_KEY=your_groq_api_key
CLIENT_URL=http://localhost:5173
HTML2PDF_API_KEY=your_html2pdf_api_key
PORT=3000

Frontend .env

Create frontend/.env with:

VITE_API_URL=http://localhost:3000

Local Development

1. Install dependencies

cd backend
npm install

cd ../frontend
npm install

2. Start the backend

cd backend
npm run dev

3. Start the frontend

cd frontend
npm run dev

4. Open the app

Frontend default URL:

http://localhost:5173

Deployment

Frontend on Netlify

• Root directory: frontend
• Build command: npm run build
• Publish directory: dist
• Environment variable:
  • VITE_API_URL=https://your-render-backend-url.onrender.com

Backend on Render

• Root directory: backend
• Build command: npm install
• Start command: npm start
• Environment variables:
  • MONGO_URI
  • JWT_SECRET
  • GOOGLE_GENAI_API_KEY
  • GROQ_API_KEY
  • HTML2PDF_API_KEY
  • CLIENT_URL=https://your-netlify-site.netlify.app
  • PORT=10000 or Render default

Cross-origin cookie setup

Because the app uses cookie-based auth across Netlify and Render:

• backend CORS must allow the Netlify frontend URL
• frontend requests must use withCredentials: true
• cookies are configured as httpOnly, secure, and sameSite: "none"

Data Models

User

• username
• email
• password

Interview Report

• title
• jobDescription
• resume
• selfDescription
• matchScore
• technicalQuestions[]
• behavioralQuestions[]
• skillGaps[]
• preparationPlan[]
• user
• timestamps

Notes

• Resume uploads are handled in memory using Multer.
• The backend currently parses resume content from PDF files.
• Groq is used first for faster responses; Gemini is used as a fallback.
• The repository still contains the default Vite README inside frontend/README.md.

Important Security Note

Real secrets are present in the current local backend/.env. Those values should not be shared publicly, and if they have been exposed anywhere outside your machine, they should be rotated.