Skip to content
Jenil1905Public

About

AI-powered interview application

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

32 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 

Repository files navigation

🎯 Intervue — AI-Powered Technical Interview Platform

React Node.js Express MongoDB Gemini AI Tailwind CSS

Intervue is an autonomous, full-stack AI mock interviewing platform designed for Computer Science students and software engineers. It simulates realistic technical interviews using real-time speech recognition, voice synthesis, code execution analysis, and proctoring controls.


🌟 Key Features

🤖 Adaptive AI Technical Interviewer

  • Topic-Tailored Questions: Dynamic generation of CS questions across Data Structures & Algorithms, Operating Systems, Computer Networks, JavaScript, Java, Python, C++, and DBMS.
  • Contextual Conversations: Analyzes both spoken explanations and written code simultaneously to ask intelligent follow-up questions.
  • Natural Intent Detection: Understands candidate intents such as requesting question repetition (repeat), seeking clarification (elaborate/explain), or requesting to move forward.

💻 Integrated Monaco Code Editor

  • Multi-Language Support: Write and practice solutions in Java, Python, C++, JavaScript, or SQL.
  • Auto-Save & Real-Time Sync: Code changes are continuously saved to MongoDB.
  • Monaco Integration: Built-in syntax highlighting, line numbers, and autocomplete powered by VS Code's editor engine.

🎙️ Real-Time Voice & Speech Interaction

  • Speech-to-Text: Converts candidate speech to text using the Web Speech API.
  • Text-to-Speech: AI interviewer speaks follow-up questions and conversational responses.
  • Streamlined Mic Controls: 1-click toggle to speak, review, and auto-submit.

🛡️ Smart Proctoring & Integrity System

  • Fullscreen Enforcement: Enforces fullscreen mode with immediate warning triggers.
  • Tab-Switch Detection: Monitors tab-switching activities with a 3-warning automated termination rule.
  • Proctoring Telemetry: Logs violation metrics directly into session reports.

📊 In-Depth Performance Analytics

  • 20-Point Metric Scoring: Questions are evaluated out of 20 points across 4 distinct categories:
    • Technical Accuracy (8 pts)
    • Code Quality & Implementation (6 pts)
    • Communication & Explanation (4 pts)
    • Problem-Solving Approach (2 pts)
  • Automated Grade Assignment: Maps overall scores (0–100) to Letter Grades (A+, A, B+, B, C+, C, F).
  • Detailed Question Breakdown: Itemized feedback, candidate strengths, improvement areas, and career recommendations.

🛠️ Architecture & Tech Stack

Layer Technologies Used
Frontend React 18, Vite, React Router v6, Tailwind CSS, Monaco Editor (@monaco-editor/react), Lucide Icons, React Calendar, Lottie React
Backend Node.js, Express.js, Mongoose (MongoDB ODM), Cookie Parser, CORS, Nodemailer, Node-cron
AI & LLM Integration Google Generative AI SDK (@google/generative-ai) — Gemini 3.6 & 2.5 Flash models
Browser APIs Web Speech API (SpeechRecognition & SpeechSynthesis), MediaDevices API (Camera/Mic)

📁 Repository Structure

Intervue/
├── client/                     # React Frontend Application
│   └── Intervue/
│       ├── public/             # Static Assets
│       ├── src/
│       │   ├── apiCalls/       # Centralized Axios API Clients
│       │   ├── assets/         # Lottie & Image Assets
│       │   ├── components/     # UI Pages & Components
│       │   │   ├── Dashboard/  # User Dashboard & Stats
│       │   │   ├── Feedback/   # Analytics & Score Reports
│       │   │   ├── InterviewPage/# Live AI Interview Room
│       │   │   ├── Login/      # Authentication Pages
│       │   │   ├── Profile/    # User Profile Settings
│       │   │   ├── Settings/   # System & AI Preferences
│       │   │   └── SignUp/     # Account Registration
│       │   ├── App.jsx         # Client Routes
│       │   └── main.jsx        # Entrypoint
│       ├── package.json
│       └── vite.config.js
└── server/                     # Express Backend Server
    ├── controllers/            # Route Request Handlers
    ├── middlewares/            # JWT Auth & Security Middleware
    ├── models/                 # Mongoose Schemas (User, Interview, Schedule)
    ├── routes/                 # API Endpoint Routers
    ├── service/                # AI Prompt Engineering & Feedback Services
    ├── connection.js           # MongoDB Database Connection
    ├── index.js                # Server Entrypoint & Cron Jobs
    └── package.json

⚙️ Environment Variables Configuration

Create a .env file inside the server/ directory with the following keys:

# Server Configuration
PORT=3000
NODE_ENV=development

# Database
dbUrl=mongodb+srv://<username>:<password>@cluster.mongodb.net/intervue

# Authentication
JWT_SECRET=your_jwt_secret_key_here

# Google Gemini AI API Key
gemini_api_key=your_google_gemini_api_key

# Email Notification Service (Gmail App Password)
EMAIL_USER=your_email@gmail.com
EMAIL_APP_PASSWORD=your_gmail_app_password

🚀 Getting Started

1. Prerequisites

  • Node.js: v18.x or higher
  • npm: v9.x or higher
  • MongoDB: Atlas Cluster or Local instance
  • Browser: Google Chrome or Chromium-based browser (for Web Speech API support)

2. Backend Setup

# Navigate to server directory
cd server

# Install dependencies
npm install

# Start development server
npm start

The backend server will run on http://localhost:3000.

3. Frontend Setup

# Navigate to client application
cd client/Intervue

# Install dependencies
npm install

# Start Vite dev server
npm run dev

Open http://localhost:5173 in your browser.


🔌 API Endpoints Summary

🔐 Authentication & User Routes (/api/auth, /api/user)

Method Endpoint Description
POST /api/auth/signup Register a new candidate account
POST /api/auth/login Authenticate and obtain JWT token
POST /api/auth/logout Clear user session cookies
GET /api/user/profile Fetch active user profile data
PATCH /api/user/profile Update profile information
DELETE /api/user/account Permanently delete user account

🎙️ Interview Management (/api/interview)

Method Endpoint Description
POST /api/interview/start/:topic Initialize a new AI interview session
GET /api/interview/session/:interviewId Fetch existing interview session details
GET /api/interview/get-history List all past interviews for user
PATCH /api/interview/save-code Auto-save code editor contents
POST /api/interview/generate-contextual-response Evaluate speech & code for AI response
POST /api/interview/get-next-question Move to next question in session
POST /api/interview/finalize-question Complete current question evaluation

📊 Feedback Routes (/api/feedback)

Method Endpoint Description
GET /api/feedback/:interviewId Retrieve saved performance feedback
POST /api/feedback/generate/:interviewId Generate dynamic Gemini feedback report

📈 Evaluation & Grading Rubric

Each completed interview session is evaluated out of 100 points (20 points × 5 questions):

Category Points Evaluation Focus
Technical Accuracy 8 pts Correctness of theoretical concepts, edge cases, and algorithm logic
Code Quality 6 pts Syntax correctness, efficiency, data structure choices, and clean code principles
Communication 4 pts Clarity of verbal explanation, structured thought process, and technical depth
Problem Solving 2 pts Methodical approach, trade-off analysis, and adaptability

📄 License

Distributed under the MIT License. See LICENSE for more information.

About

AI-powered interview application

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages