🎙️ AI Interview Simulator

A full-stack AI-powered mock interview platform that helps job seekers practise technical and behavioural interviews with real-time webcam analysis, AI-generated questions, and detailed post-session reports.

---

✨ Features

• 🤖 AI-Generated Questions — Dynamic interview questions tailored to job role and level, powered by OpenAI / Groq
• 🎯 Real-Time Webcam Analysis — Eye contact and head stability tracking using MediaPipe FaceMesh (runs locally, no video stored)
• 🎤 Speech-to-Text Answer Input — Speak or type your answers freely
• 📊 Post-Interview Report — Detailed scoring with feedback per question and downloadable PDF
• 🔐 JWT Authentication — Secure access & refresh token flow with Redis token blacklisting
• 📈 Dashboard & History — View past sessions and analytics over time

---

🛠️ Tech Stack

Frontend (/client)

Tech	Purpose
React 19 + TypeScript	UI framework
Vite 8	Build tool & dev server
TailwindCSS v4	Styling
shadcn/ui	Component library
Zustand	Global auth state
TanStack Query	Server state & caching
React Hook Form + Zod	Form validation
MediaPipe FaceMesh	Local face tracking
Recharts	Analytics charts
React Router v7	Client-side routing

Backend (/server)

Tech	Purpose
Node.js + Express 5	REST API server
MongoDB + Mongoose	Primary database
Redis (Upstash)	Refresh token store / blacklist
OpenAI / Groq SDK	AI question & feedback generation
JWT (jsonwebtoken)	Authentication
bcrypt	Password hashing
PDFKit	PDF report generation
Zod	Request validation
Helmet + Rate Limit	Security hardening

---

📁 Project Structure

AI-Interview/
├── client/                  # React frontend
│   ├── src/
│   │   ├── components/      # Reusable UI components
│   │   ├── hooks/
│   │   │   ├── auth/        # useLogin, useLogout, useRefreshToken
│   │   │   ├── data/        # useAnalytics, useReport, useInterviewHistory
│   │   │   ├── interview/   # useInterview, useStartInterview, useSubmitAnswer
│   │   │   └── vision/      # useMediaPipe, useWebcam, useEyeContact, useHeadPose
│   │   ├── pages/           # LoginPage, RegisterPage, Dashboard, InterviewPage, ReportPage
│   │   ├── services/        # Axios instance (with interceptors)
│   │   └── store/           # Zustand auth store
│   └── index.html
│
└── server/                  # Express backend
    └── src/
        ├── config/          # MongoDB & Redis connection
        ├── controllers/     # auth, interview, answer, camera, report
        ├── middleware/      # JWT auth, Zod validation
        ├── models/          # User, Interview, Question, Answer, Report, CameraAnalysis
        ├── routes/          # auth, interview, camera, report, user
        ├── services/        # AI service, report generation
        └── utils/           # Zod validation schemas

---

🚀 Getting Started

Prerequisites

• Node.js ≥ 20
• MongoDB Atlas account
• Redis instance (e.g. Upstash)
• OpenAI or Groq API key

1. Clone the repo

git clone https://github.com/your-username/AI-Interview.git
cd AI-Interview

2. Setup the backend

cd server
cp .env.example .env    # Fill in your credentials
npm install
npm run dev             # Runs on http://localhost:5000

3. Setup the frontend

cd client
cp .env.example .env    # Set VITE_API_URL
npm install
npm run dev             # Runs on http://localhost:5173

---

🔑 Environment Variables

server/.env

NODE_ENV=development
PORT=5000
MONGODB_URI=mongodb+srv://<user>:<pass>@cluster.mongodb.net/ai-interview
JWT_ACCESS_SECRET=your_access_secret_min_32_chars
JWT_REFRESH_SECRET=your_refresh_secret_min_32_chars
JWT_ACCESS_EXPIRES=15m
JWT_REFRESH_EXPIRES=7d
REDIS_URL=redis://:password@your-upstash-url:6379
OPENAI_API_KEY=sk-...
ALLOWED_ORIGINS=http://localhost:5173

client/.env

VITE_API_URL=http://localhost:5000/api

---

📡 API Endpoints

Method	Endpoint	Description
POST	/api/auth/register	Register new user
POST	/api/auth/login	Login & get tokens
POST	/api/auth/refresh	Refresh access token
POST	/api/auth/logout	Logout & blacklist token
POST	/api/interviews/start	Start new interview session
POST	/api/interviews/:id/submit	Submit answer for a question
GET	/api/interviews	Get interview history
GET	/api/report/:id	Get full interview report
GET	/api/report/:id/pdf	Download report as PDF
POST	/api/camera/analysis	Save camera analysis data

---

🎥 Webcam Privacy

All face analysis is done entirely in the browser using MediaPipe FaceMesh.
No video frames or images are ever sent to the server.
Only computed metrics (eye contact %, head stability %) are stored.

---

📄 License

MIT — feel free to use, fork, and improve.