AI-Powered Job Preparation Platform

A production-ready full-stack GenAI web application that helps users prepare for job interviews by analyzing their resume against job descriptions, detecting skill gaps, generating personalized interview questions, and creating ATS-optimized resumes.

Work in Progress — BUG in gemini API integration.

---

Features

• Secure Authentication — JWT with token blacklisting, OTP email verification, httpOnly cookies
• Resume Upload & Parsing — Upload resume and job description for AI analysis
• Skill Gap Detection — Gemini AI detects missing skills based on JD vs resume
• Interview Question Generation — Role-specific technical and behavioural questions
• Personalized Prep Plan — Structured preparation plan based on detected gaps
• ATS Resume Generation — Puppeteer-based dynamic PDF resume tailored to the job

---

Tech Stack

Frontend

Tech	Purpose
React.js + Vite	UI framework
React Router	Client-side routing
Axios	HTTP requests
Context API	Global state management
SCSS	Styling

Backend

Tech	Purpose
Node.js + Express.js	Server & API
MongoDB + Mongoose	Database
JWT	Authentication
bcrypt	Password hashing
cookie-parser	Cookie management
Gemini AI	AI features
Puppeteer	PDF generation

---

Project Structure

GenAI/
├── Backend/
│   ├── src/
│   │   ├── controllers/                 # auth.controller.js , interview.controller.js
│   │   ├── models/                      # user.model.js, blacklist.model.js, report.model.js
│   │   ├── routes/                      # auth.route.js, interview.route.js
│   │   ├── middlewares/                 # auth.middleware.js, file.middleware.js 
│   │   ├── config/                      # database.js
│   │   ├── services/                    # ai.service.js
│   │   └── app.js
│   ├── .env                             # Never committed
│   ├── .gitignore
│   ├── package.json
│   └── server.js
├── Frontend/
│   ├── src/
│   │   ├── features/
│   │   │   ├── auth/
│   │   │   │   ├── components           # Protected.jsx
│   │   │   │   ├── hooks/               # useAuth.jsx
│   │   │   │   ├── pages/               # Login.jsx, Register.jsx
│   │   │   │   ├── styles/              # auth.form.scss
│   │   │   │   ├── services/            # auth.api.jsx
│   │   │   │   └── auth.context.jsx
│   │   │   └── interview/
│   │   │       ├── components           
│   │   │       ├── hooks/               # useInterview.jsx
│   │   │       ├── pages/               # Home.jsx, Interview.jsx
│   │   │       ├── styles/              # home.scss, interview.scss
│   │   │       ├── services/            # interview.api.jsx
│   │   │       └── interview.context.jsx
│   │   ├── styles/                      # button.scss
│   │   ├── App.jsx
│   │   ├── style.scss
│   │   └── main.jsx
│   ├── index.html
│   └── package.json
└── README.md

---

Getting Started

Prerequisites

• Node.js v18+
• MongoDB (local or Atlas)
• Google Gemini API Key

1. Clone the repository

git clone https://github.com/Ri1tik/GenAI.git
cd GenAI

2. Setup Backend

cd Backend
npm install

Create .env file in Backend/:

PORT=3000
MONGO_URI=your_mongodb_connection_string
JWT_SECRET=your_jwt_secret_min_32_chars
GOOGLE_API_KEY=your_gemini_api_key
NODE_ENV=development

Start the backend:

node server.js
# or with auto-restart
nodemon server.js

3. Setup Frontend

cd Frontend
npm install

Create .env file in Frontend/:

VITE_API_URL=http://localhost:3000

Start the frontend:

npm run dev

4. Open in browser

http://localhost:5173

---

API Endpoints

Auth Routes — /api/auth

Method	Endpoint	Access	Description
POST	/register	Public	Register a new user
POST	/login	Public	Login a user
GET	/logout	Public	Clear auth token from cookies and blacklist it
GET	/getMe	Private	Get details of the currently authenticated user

Interview Routes — /api/interview

Method	Endpoint	Access	Description
POST	/	Private	Generate interview report based on candidate's resume, self description, and job description
GET	/:interviewId	Private	Get interview report by ID
GET	/	Private	Get all interview reports of the logged-in user
GET	/resume/pdf/:interviewReportId	Private	Download AI-generated resume PDF for the interview report

---

Security Features

• JWT stored in httpOnly cookies (not localStorage)
• Token blacklisting on logout prevents session reuse
• Passwords hashed with bcrypt (salt rounds: 10)
• Protected routes via Express middleware
• CORS configured with credentials: true
• Environment variables validated on startup

---

Author

Ritik Roshan Yadav

• B.Tech CSE — NIT Rourkela (CGPA: 9.03)
• GitHub: @Ri1tik
• ritik.r.yadav0001@gmail.com

---

License

This project is for educational and portfolio purposes.