InterviewAI

InterviewAI is a full-stack web application that helps candidates prepare for interviews using AI.

Users can upload a resume, add their self-description and a target job description, and receive:

• A job match score
• Technical and behavioral interview questions with guidance
• Skill gap analysis
• A day-wise preparation plan
• A tailored resume PDF

Features

• Cookie-based authentication with JWT
• Secure protected routes
• Resume upload and text extraction from PDF
• AI-generated interview preparation report
• Report history and detailed report pages
• AI-generated, downloadable resume PDF
• Neumorphism-inspired frontend design system

Why You Should Use InterviewAI

• Clear, structured feedback instead of generic chat answers
• Practice sessions with scoring to reveal weak dimensions
• Actionable next steps via skill gaps, prep plans, and roadmaps
• Resume-to-report workflow reduces setup time
• End-to-end flow from practice to improvement plan

How To Get Maximum Benefit

• Use a specific job description to sharpen the report context
• Keep your self-description honest and focused on real experience
• Run mock sessions weekly and compare trends over time
• Review the question bank and rate yourself right after each session
• Generate roadmaps from your latest report and complete tasks in order
• Revisit reports after every major resume or skill update

Concepts Used In This Project

• AI-first workflows with JSON schema validation for stable outputs
• Idempotent session flows to safely handle retries and disconnects
• Spaced repetition scheduling for question bank reviews
• Adaptive follow-up questions based on weak score dimensions
• Role-based report generation from resume and job description
• Neumorphic UI design system for consistent visuals
• Cookie-based auth with JWT and token blacklisting

Tech Stack

Frontend

• React 19
• Vite 7
• React Router DOM
• Redux Toolkit + Redux Persist
• Tailwind CSS 4

Backend

• Node.js + Express 5
• MongoDB + Mongoose
• Google GenAI SDK (Gemini)
• Puppeteer (PDF generation)
• Multer (file upload)
• bcryptjs + jsonwebtoken

Monorepo Structure

interviewAI/
  backend/
    server.js
    package.json
    src/
      app.js
      config/
      controllers/
      middleware/
      model/
      routes/
      service/
  frontend/
    package.json
    src/
      components/
      pages/
      services/
      store/
  docs/
    api-overview.md
    controllers/

Prerequisites

• Node.js 20+ recommended
• npm 10+ recommended
• MongoDB (local or Atlas)
• Gemini API key

Environment Variables

Create a .env file inside backend/.

MONGO_URI=your_mongodb_connection_string
JWT_SECRET=your_jwt_secret
GEMINI_API_KEY=your_gemini_api_key

Optional frontend env in frontend/.env:

VITE_API_BASE_URL=http://localhost:3000

Local Setup

1) Install dependencies

From the project root:

cd backend
npm install

cd ../frontend
npm install

2) Start backend

In one terminal:

cd backend
npm run dev

Backend runs on http://localhost:3000.

3) Start frontend

In a second terminal:

cd frontend
npm run dev

Frontend runs on http://localhost:5173.

Scripts

Backend (backend/package.json)

• npm run dev - Run backend with nodemon
• npm test - Placeholder test script (currently not implemented)

Frontend (frontend/package.json)

• npm run dev - Start Vite dev server
• npm run build - Build production bundle
• npm run preview - Preview production build
• npm run lint - Run ESLint

Authentication Model

• Auth uses an HTTP cookie named token
• JWT token is generated on register/login
• Logout clears cookie and stores token in a blacklist collection
• Protected routes validate token and blacklist status

Frontend requests must include credentials:

• Fetch: credentials: "include"

API Overview

Base URL: http://localhost:3000

Auth

• POST /api/auth/register - Register user
• POST /api/auth/login - Login user
• GET /api/auth/logout - Logout user
• GET /api/auth/get-me - Get current logged-in user

Interview Reports

• POST /api/interview - Create interview report
• GET /api/interview - Get all reports for logged-in user
• GET /api/interview/report/:interviewId - Get one report by id
• GET /api/interview/resume/pdf/:interviewReportId - Download generated resume PDF

Report Creation Request

POST /api/interview expects multipart/form-data:

• resume (PDF file, max 3 MB)
• selfDescription (string)
• jobDescription (string)

Data Models (Summary)

User

• username (unique)
• email (unique)
• password (hashed)

InterviewReport

• title
• jobDescription
• resume
• selfDescription
• matchScore (0-100)
• technicalQuestions[]
• behavioralQuestions[]
• skillGaps[]
• preparationPlan[]
• user (ObjectId)
• timestamps

Blacklist Token

• token
• timestamps

Application Flow

1. User registers or logs in
2. User creates a new report from dashboard
3. User uploads resume + self-description + job description
4. Backend parses resume PDF and calls Gemini
5. AI response is validated and saved in MongoDB
6. User views detailed report
7. User can download tailored resume as PDF

CORS and Cookie Notes

• Backend CORS origin is set to http://localhost:5173
• Cookie auth requires frontend and backend to run on configured local ports
• Cross-origin requests must include credentials

Troubleshooting

Mongo connection fails

• Check MONGO_URI
• Ensure MongoDB service/Atlas cluster is reachable

401 on protected routes

• Make sure frontend sends credentials
• Login again to refresh cookie
• Confirm token has not been blacklisted after logout

Resume upload fails

• Confirm file is a PDF
• Confirm file size is under 3 MB

Gemini-related failures

• Check GEMINI_API_KEY
• Ensure API key has access to Gemini model used in the backend

Known Limitations

• No automated tests configured yet
• Limited central error handling around AI calls
• No queue/retry mechanism for long AI/PDF operations
• No rate limiting on auth/report endpoints

Documentation

Additional docs are available in:

• docs/api-overview.md
• docs/controllers/auth.controllers.md

Contributing

1. Fork the repo
2. Create a feature branch
3. Make focused changes
4. Run frontend lint and verify backend startup
5. Open a pull request with clear description

Roadmap Ideas

• Add test coverage (unit + integration)
• Add request validation and global error handling
• Add rate limiting and hardening for production
• Add report generation jobs/queue for scale
• Add password reset and email verification

License

Copyright (c) 2026 Manthan Licensed under the MIT License. You are free to use, copy, and modify this software, provided this notice is included.