Intervyo – AI-Powered Interview Simulation Platform

Intervyo is an AI-driven interview preparation and evaluation platform designed to simulate real-world technical and HR interviews.
It helps candidates practice interviews, receive structured, criteria-based feedback, and improve performance through AI analysis instead of vague human opinions.

This is not a generic “chat with AI” project.
Intervyo is built for realism, accountability, and measurable improvement.

📚 Table of Contents

• Why Intervyo Exists
• Core Features
  • AI Interview Simulation
  • Smart Evaluation & Feedback
  • Interview History & Progress Tracking
• Advanced Multi-Company Features (NEW)
  • Smart Company Recommendation Engine
  • Company Interview Calendar Integration
  • Real Interview Question Database
• Tech Stack
• System Architecture
• Installation & Setup
  • Prerequisites
  • Clone the Repository
  • Backend Setup
  • Frontend Setup
  • Environment Variables
• Docker (Optional)
• Current Status
• Use Cases
• Design Philosophy
• Contributing
• Code of Conduct

---

🎯 Why Intervyo Exists

Most interview preparation platforms fail because they:

• Ask generic questions
• Give fluffy, non-actionable feedback
• Do not simulate real interview pressure

Intervyo fixes this by:

• Running structured interviews
• Evaluating responses against defined criteria
• Giving actionable feedback, not motivational nonsense

If it doesn’t help you perform better in a real interview, it doesn’t belong here.

---

🧠 Core Features

🎤 AI Interview Simulation

• Technical, behavioral, and mixed interview modes
• Timed questions to simulate real interview pressure
• Adaptive follow-up questions based on candidate responses
• New: Real-time Body Language Coach (Eye contact & Posture tracking) 👁️

📊 Smart Evaluation & Feedback

• Communication clarity analysis
• Technical correctness scoring
• Confidence & structure assessment
• Strengths, weaknesses, and improvement suggestions
• Live Confidence HUD during interviews 💯

📁 Interview History & Progress Tracking

• Store past interviews
• Compare performance over time
• Identify recurring weaknesses

🔐 Secure User System

• Authentication & authorization
• Private interview data
• Secure API handling

---

🚀 Advanced Multi-Company Features (NEW)

🤖 Smart Company Recommendation Engine

• AI-powered analysis of your interview performance
• Personalized company recommendations based on skill level
• Success probability calculation for each company
• Gap analysis with improvement suggestions
• Route: /advanced-features or /api/recommendations

📅 Company Interview Calendar Integration

• Track upcoming interview dates with countdown timers
• Automatically generated preparation milestones
• Daily practice recommendations based on days remaining
• Progress tracking and readiness score
• Route: /api/calendar

💎 Real Interview Question Database

• Crowdsourced real interview questions from actual interviews
• Voting system (upvote/downvote) for question quality
• Question verification workflow
• Frequency tracking (how often questions are asked)
• Search and filter by company, difficulty, type
• Trending questions feature
• Route: /api/questions

🤝 Interview Buddy Matching

• Find compatible study partners preparing for same companies
• Compatibility algorithm based on target companies and skill level
• 1-on-1 buddy connections with mock interview scheduling
• Study group creation and management
• Route: /api/buddy

📊 Company-Specific Success Metrics

• Enhanced company profiles with hiring bar benchmarks
• Success thresholds for each interview type
• Difficulty ratings and acceptance rates
• Historical performance statistics

🎤 Speech Practice Lab (Frontend)

• Real-time speech-to-text using Web Speech API
• Live metrics: words, WPM, average sentence length, filler words
• Coaching tips for pace and clarity
• Save sessions locally for quick review (no backend required)
• Route: /practice-lab
• Requires microphone permission in the browser (Chrome recommended)

🎬 Interview Replay System (NEW)

• Full Playback: Review completed interviews with complete conversation history
• Timestamped Notes: Add personal notes at any point with categorization (improvement, strength, mistake, learning)
• Smart Bookmarks: Quick-jump to important moments in the interview
• Resume Functionality: Pick up where you left off during review sessions
• Global Search: Search across all notes and bookmarks from all interviews
• View Analytics: Track how often you review each interview and total watch time
• Secure Sharing: Generate share links to get feedback from mentors or study buddies
• Self-Reflection: Identify patterns and track improvement over time
• Route: /api/replay
• Perfect for: Post-interview analysis, mentor feedback, peer review, progress tracking

🎯 AI-Powered Weakness Predictor & Attack Plan (NEW - OUT OF BOX!)

• Predictive Intelligence: Analyzes your last 20 interviews to predict where you'll fail BEFORE your next interview
• Personalized Attack Plans: 3-phase improvement strategy (Emergency Fixes → Strengthen Core → Polish & Perfect)
• Micro-Challenges: 15 bite-sized, actionable tasks targeting your specific weaknesses (30-90 min each)
• Success Probability: Get real probability scores for easy/medium/hard interviews and specific companies
• Real-Time Progress Tracking: Improvement score, completion percentage, trend analysis (improving/declining/stable)
• AI Insights: Hidden strengths, blind spots, quick wins, peer comparison, long-term goals
• Weakness Categories: Tracks 10 areas (technical-depth, system-design, coding-efficiency, communication-clarity, etc.)
• Severity Levels: Critical (urgent), High (significant), Medium (polish needed), Low (strengths)
• Route: /api/attack-plan
• Unique value: Proactive vs Reactive - Know your failure points before they happen, not after

---

🛠 Tech Stack

🎨 Frontend

• React
• Tailwind CSS
• Responsive UI (desktop + mobile)

⚙️ Backend

• Node.js
• Express.js
• MongoDB
• REST APIs

🤖 AI Layer

• LLM-based interview logic
• Prompt-engineered evaluation criteria
• Structured scoring system (not random text output)

---

🧩 System Architecture (High Level)

flowchart LR
U[User]
F[Frontend - React]
B[Backend API - Express]
AI[AI Evaluation Engine]
DB[(MongoDB Database)]
AN[Feedback & Analytics]

U --> F
F --> B
B --> AI
AI --> DB
DB --> AN
AN --> F

Loading

Simple, scalable, and not over-engineered.

---

⚙️ Installation & Setup

📦 Prerequisites

• Node.js (v18+ recommended)
• MongoDB
• Git

---

📥 Clone the Repository

git clone https://github.com/santanu-atta03/Intervyo  
cd intervyo

---

🔧 Backend Setup

cd backend  
npm install  
npm run dev  

---

⚠️ React Version Compatibility Note

This project currently uses React 19.

Some dependencies do not yet officially support React 19.
As a result, running npm install may fail with an ERESOLVE peer dependency error.

Temporary Workaround

Until full React 19 support is available across dependencies, install frontend packages using:

npm install --legacy-peer-deps

---

💻 Frontend Setup

cd frontend  
npm install  
npm run dev

---

🔑 Environment Variables

Create a .env file in the backend directory:

PORT=5000  
MONGO_URI=your_mongodb_connection_string  
AI_API_KEY=your_ai_api_key  

---

Docker (Optional)

This setup is for local development only and does not change the default workflows.

1. Create any needed backend env values (optional). The Docker Compose config uses Backend/.env.example by default and overrides the MongoDB host.

2. Start the stack:

docker compose up --build

Frontend: http://localhost:5173 Backend: http://localhost:5000

If you want to point the frontend to a different API URL, set VITE_API_BASE_URL before building.

---

For a deeper walkthrough and rationale, see docker_guide.md.

🚦 Current Status

• Core interview flow implemented
• AI-based evaluation logic working
• User authentication
• Advanced analytics (in progress)
• Multi-role interview templates (planned)

---

🎯 Use Cases

• Students preparing for placements
• Developers preparing for technical interviews
• Self-assessment before real interviews
• Mock interview practice without human bias

---

🧠 Design Philosophy

• Realism over gimmicks
• Feedback over praise
• Skill improvement over vanity metrics

This platform is built to expose weaknesses, not hide them.

---

🤝 Contributing

Please read CONTRIBUTING before opening a pull request.
Low-effort, spam, or cosmetic-only contributions will be closed.

---

📜 Code of Conduct

This project follows the Contributor Covenant Code of Conduct.
Please read CODE_OF_CONDUCT before contributing.