An AI-powered interview simulation platform that allows users to practice technical interviews with dynamic question flow, real-time response evaluation, and structured performance feedback.
-
🎯 Dynamic Interview Flow Simulates real interview scenarios with multiple questions and structured progression
-
🧠 AI-Based Answer Evaluation Evaluates user responses and generates:
- Score (0–10)
- Strengths & weaknesses
- Improvement suggestions
-
📊 Performance Report Dashboard Displays:
- Overall score
- Per-question analysis
- Feedback summary
-
🔐 Authentication & Session Management Secure login system using JWT with protected routes
-
☁️ Production Deployment Hosted on AWS EC2 with Nginx reverse proxy and PM2 process management
Frontend
- React.js (Vite)
- Tailwind CSS
Backend
- Node.js
- Express.js
Database
- MongoDB (Mongoose)
Authentication
- JWT (JSON Web Tokens)
AI Integration
- API-based response evaluation (LLM-powered)
DevOps & Deployment
- AWS EC2
- Nginx
- PM2
Client (React)
↓
Backend API (Express)
↓
Interview Service Layer
├── Question Flow Controller
├── Answer Evaluation Engine
├── Feedback Generator
↓
Database (MongoDB)
git clone https://github.com/your-username/ai-interviewer.git
cd ai-interviewer
cd backend
npm install
Create .env file:
PORT=5000
MONGO_URI=your_mongodb_connection
JWT_SECRET=your_secret
AI_API_KEY=your_api_key
Run server:
npm run dev
cd frontend
npm install
npm run dev
- Backend deployed on AWS EC2
- Reverse proxy configured using Nginx
- Process managed with PM2
- HTTPS enabled via Certbot
- Voice-based interview (Speech-to-Text / Text-to-Speech)
- Advanced answer scoring using embeddings
- Real-time interview mode
- Interview analytics dashboard
- GitHub: https://github.com/Aditya-2003
- LinkedIn: https://www.linkedin.com/in/aditya-shrivas-29b111256
This project uses AI-based APIs for response evaluation and does not train custom machine learning models.