Intbar

A full-stack interview system designed to simulate real-time technical interviews using AI, code execution APIs, and proctoring features. The platform evaluates candidates through coding, voice interaction, and behavioral monitoring.

---

Overview

This project is built to replicate a real interview environment where users can attempt coding questions, receive AI-driven interactions, and be monitored for fairness. It integrates external APIs and real-time analysis to provide a comprehensive evaluation system.

---

How the Project Works

1. User Interaction (Frontend)

   • Users access the platform and participate in interview sessions.
   • Questions are displayed dynamically based on difficulty levels.
   • A timer is assigned per question (easy, medium, hard).

2. Code Execution (Judge0 API)

   • Users write code directly in the platform.
   • Code is sent to the Judge0 API for execution.
   • Output is returned and displayed instantly based on user input.

3. AI Interaction (Grok API)

   • AI is used to simulate interview questions and responses.
   • It can generate follow-up questions and evaluate user inputs.
   • Helps create a dynamic and adaptive interview experience.

4. Real-Time Camera Analysis

   • The system accesses the user's camera during the interview.
   • It monitors user behavior to ensure fairness.
   • Detects suspicious activities such as absence or multiple faces.

5. Voice Recognition System

   • Captures user voice responses during the interview.
   • Converts speech to text for processing and evaluation.
   • Enables a more realistic interview environment.

6. Backend Processing

   • Handles API requests, business logic, and validations.
   • Manages interview sessions, question flow, and timing.
   • Communicates with external APIs (Judge0 and Grok).

7. Database Management

   • Stores user data, interview records, and results.
   • Maintains question banks and session logs.

---

Tech Stack

• Frontend: React
• Backend: Node.js with Express
• Database: MongoDB
• Code Execution: Judge0 API
• AI Integration: Grok API
• Additional: WebRTC / Media APIs for camera and voice processing

---

Key Features

• Real-time coding interview environment
• Integration with Judge0 API for code execution
• AI-powered interview interaction using Grok API
• Live camera monitoring for proctoring
• Voice recognition for verbal responses
• Timed questions based on difficulty level
• Secure and scalable architecture

---

Project Structure

• /frontend – User interface and interaction logic
• /backend – API handling and server logic
• /routes – API endpoints
• /models – Database schemas
• /controllers – Core logic and processing
• /config – Environment and database configuration

---

Setup Instructions

1. Clone the repository git clone

2. Install dependencies

   • Backend: npm install
   • Frontend: npm install

3. Configure environment variables

   • Add API keys for Judge0 and Grok
   • Add database connection string

4. Run the application

   • Backend: npm start
   • Frontend: npm start

5. Open in browser http://localhost:3000

---

Key Highlights

• Combines AI, real-time systems, and full-stack development
• Demonstrates API integration with external services
• Implements real-time monitoring and interaction
• Designed to simulate real-world interview scenarios

---

Future Improvements

• Advanced AI evaluation and scoring system
• Cheating detection using AI models
• Detailed performance analytics dashboard
• Multi-language support for coding interviews

---

Conclusion

This project demonstrates the ability to build an advanced, real-time full-stack system integrating AI, third-party APIs, and interactive features. It reflects strong problem-solving skills and practical experience in developing scalable and intelligent applications.