🎯 InterviewAI - Professional Interview Intelligence Platform

Created & Authored by: Karan Shelar

📚 Project Docs: Challenges | Next Steps | Coding Round Design | Technical Flow | Collaborators | All Docs →

🌟 The Simulation Philosophy

InterviewAI is not just a "mock interview" tool; it is an Advanced Intelligent Simulation Entity. While traditional platforms rely on static question banks, InterviewAI uses a Dynamic Agentic Loop to synthesize the reality of the 2025/2026 job market.

🎭 The Persona Architecture

We don't just ask questions; we simulate Human Personalities:

Adinath (The Primal Sage): Simulates the "Foundation First" interviewer. He is cold, direct, and explores the recursive depth of your technical architecture. He tests the "How" and the "Why."
Veda (The Eternal Wisdom): Simulates the "Clarity Specialist." She is observant, testing your vision, cultural alignment, and your ability to communicate complex ideas under pressure.

🚀 Technical Innovation: The "Agentic Edge" (v2.3 — Domain-Aware Round Engine)

What makes this project unique in the industry is its Three-Tier Intelligence Architecture, designed to solve the "Knowledge Gap" in AI:

The Curated Core: Local access to 402 expert-verified company profiles across 12 domains for zero-latency, reliable data.
The Agentic Discovery Loop (LangGraph): An autonomous multi-agent team that researches public companies in real-time.
- Dynamic Domain Intelligence: Automatically prevents "Role Forcing" (e.g., won't force LeetCode questions on a legal or clinical role).
- Geographic Guardrails: Prevents acronym collisions (e.g., identifying MOC Cancer Care India vs Moffitt USA) by localized search auditing.
- Generic Article Purge: Python-level domain blocklist strips SEO junk (datacamp, guru99, intellipaat, etc.) before the AI ever sees the data.
Stealth Mode & Synthetic Fallback: A world-first feature that Reverse-Engineers company DNA from a Job Description. If a company is in "Stealth Mode" (non-public), the AI synthesizes a high-fidelity interview based on industry-standard benchmarking.
Domain-Aware Round Engine (v2.3): Each of the 12 career domains gets its own interview structure — Finance gets Case Study rounds, Healthcare gets Situational/Clinical rounds, Creative gets Portfolio Review. No more forcing tech candidates' rounds onto Accounting or Legal applicants.
Crowdsourced Self-Learning Memory (v2.4): The system now learns from its users. The Stealth Registry implements a "Witness vs Expert" consensus logic. It builds high-fidelity profiles for private startups by aggregating data from persistent "Strugglers" (Structural Roadmap) and Top-Performers (Success Criteria).

🛡️ Confidence Score Guide

The system now provides transparency on how it "trusts" the discovered intelligence:

Score	Status	Description	Reliability
0 - 20	Synthetic	No public info found. AI "guesstimates" based strictly on the Industry & JD.	⚠️ Experimental
85	Verified	Sufficient public sources found. Identity is verified.	✅ High
100 - 160	Elite Certified	Extreme certainty. Matched Location, Industry, and found Direct Interview Questions.	🏆 Production Grade

🏗️ System Architecture

graph TD
    User((User)) -->|"Uploads Resume/JD"| UI["React Dashboard"]
    UI -->|Trigger| Backend["FastAPI Core"]
    
    subgraph "The Intelligence Brain"
        Backend --> Gatekeeper{Entry Gate}
        Gatekeeper -->|Found| DB["Curated 402 DB"]
        Gatekeeper -->|"Not Found"| Agents["LangGraph Research Team"]
        
        Agents --> Router["Router (Location Guard)"]
        Router --> Researcher["DuckDuckGo Researcher"]
        Researcher --> Auditor["Auditor Agent - The Bouncer"]
        Auditor --> Architect["Architect (Llama-3 Profile Builder)"]
        Architect --> Critic["Gemini Quality Validator"]
        
        Critic -->|Approved| Memory[("(discoveries.json)")]
        Memory --> Context["Simulation Context"]
        DB --> Context
    end
    
    Context -->|Seed| Simulation["Gemini 2.0 Simulation"]
    Simulation -->|"Voice/Video"| User
    Simulation -->|Feedback| Persistence[("Neon Cloud PostgreSQL")]

✅ Development Progress

Phase 1: Core Foundation (COMPLETED ✅)

Backend Infrastructure

FastAPI application setup with CORS
PostgreSQL database integration (Neon Cloud)
SQLAlchemy ORM models
Alembic migration system (Production Sync Ready)
Gemini 2.0 Flash API integration
PDF resume parsing (PyPDF)
Session management system
RESTful API endpoints

Frontend Foundation

AI Interview System

Premium UI/UX

🚧 Roadmap - What's Next

🔥 Phase 2: Multi-Round Interview System (PENDING 🔴)

Round 1: Technical/Screening Round

Basic implementation (CURRENT)
Round-specific evaluation criteria
Technical depth assessment
Coding problem integration (optional)

Round 2: Behavioral/HR Round

Round 3: System Design Round (Technical Roles)

System design problem generation
Scalability discussion prompts
Trade-off analysis evaluation
Architecture diagram support (future)

Round 4: Managerial Round

Leadership scenario questions
Conflict resolution scenarios
Team management assessment
Strategic thinking evaluation

Round 5: Final/Director Round

Vision and long-term goals
Company culture alignment
Executive presence evaluation
Offer negotiation simulation

Non-Technical Interview Rounds

Healthcare & Medical: Clinical scenarios, patient interaction
Business & Management: Case studies, market analysis
Finance & Accounting: Financial modeling, risk assessment
Creative & Design: Portfolio review, design thinking
Sales & Marketing: Pitch simulation, objection handling
Education & Training: Teaching methodology, curriculum design
Legal: Case analysis, ethical scenarios
Hospitality & Tourism: Customer service scenarios

� Phase 3: Advanced Analytics & Scoring (PENDING 🔴)

ATS & Resume Analysis

Performance Analytics

Multi-round score aggregation
Performance trends over time
Strengths/weaknesses dashboard
Comparison with industry benchmarks
Detailed transcript analysis
Voice tone analysis (pitch, pace, clarity)
Confidence score tracking

Behavioral Analysis

Basic vibe analysis in evaluation
Hesitation pattern detection
Filler word counting (um, uh, like)
Speaking pace analysis
Assertiveness scoring
Body language feedback (if camera enabled)

🎓 Phase 4: Learning & Growth (PENDING 🔴)

Post-Interview Learning

7-Day Learning Roadmap generation
Failed Topics Identification
Resource Recommendations (courses, articles, videos)
Practice Problem Sets
Retry Interview after learning period

Skill Development

🔐 Phase 5: User Management & Authentication (IN PROGRESS 🟡)

User System

Pricing Tiers & Payment

Free Tier: 1 interview/2 weeks, basic feedback
Pro Tier (₹199): Unlimited interviews, JD-tailored questions
Elite Tier (₹499): Panel mode, 7-day roadmap, vibe analysis
UPI QR Code Payment - Direct UPI payment (no gateway fees)
Payment verification system
Manual subscription activation after payment screenshot

UPI Payment Flow (Zero Cost)

User selects Pro/Elite tier
System displays UPI QR Code with amount
User pays via any UPI app (Google Pay, PhonePe, Paytm)
User uploads payment screenshot
Admin verifies payment (manual/automated OCR)
Subscription activated instantly
No payment gateway fees - 100% of payment received

☁️ Phase 6: Deployment & Infrastructure (PENDING 🔴)

Backend Deployment (AWS)

AWS EC2 - FastAPI backend hosting
AWS Elastic IP - Static IP for backend
AWS Security Groups - Firewall configuration
Nginx - Reverse proxy setup
Gunicorn - WSGI server for FastAPI
Environment variable management
SSL/TLS certificates (Let's Encrypt - Free)
Auto-restart on failure (systemd)

Database Migration

Current: PostgreSQL (pgAdmin4 - Local development)
Production: Supabase PostgreSQL (Free tier - 500MB)
Database migration from local to Supabase
Connection pooling setup
Backup strategy (Supabase auto-backup)
Environment-based DB switching

Frontend Deployment

Vercel deployment (Free tier)
GitHub integration for auto-deploy
Environment variable setup (API URL)
Custom domain (free .vercel.app subdomain)
CDN configuration (included)
Performance optimization
Analytics setup (Vercel Analytics - Free)

DevOps Pipeline

GitHub - Source control & version management
GitHub Actions - CI/CD pipeline (Free)
Automated testing on PR
Auto-deploy to Vercel on main branch push
Backend deployment automation
Database migration scripts
Monitoring (AWS CloudWatch Free tier)
Error tracking (Sentry Free tier)

🚀 Phase 7: Advanced Features (FUTURE 🔵)

AI Enhancements

Voice cloning for more realistic interviewers
Video avatar generation
Real-time facial expression analysis
Multi-language support
Industry-specific jargon training

Collaboration Features

Integration & API

✨ Current Features (v1.0.1)

🎤 Live Interview Simulation

✅ Real-time voice recognition for natural conversation
✅ Dynamic AI responses adapting to your answers
✅ Industry-standard UI with professional glassmorphism
✅ Live video feed with camera controls
✅ Minimum 5 questions before evaluation

🧠 Intelligent Interview System

✅ Multi-persona AI interviewers (Adinath & Veda)
✅ Company Intelligence System (v2.0)
- ✅ Curated database for 383 companies (20+ per major domain)
- ✅ Engineering & Tech: Google, Amazon, Microsoft, Nvidia, Tesla, OpenAI (134 total)
- ✅ Healthcare & Medical: Pfizer, Mayo Clinic, Johnson & Johnson, Pfizer (22 total)
- ✅ Finance & Accounting: Goldman Sachs, JPMorgan, Stripe, Coinbase (22 total)
- ✅ Legal: Kirkland & Ellis, DLA Piper, Latham & Watkins, Clio (22 total)
- ✅ Construction & Trades: AECOM, Caterpillar, Larsen & Toubro, Vinci (23 total)
- ✅ Creative & Design: Pixar, DreamWorks, Adobe, Vimeo, SoundCloud (21 total)
- ✅ Education & Training: Coursera, Udacity, Khan Academy, Duolingo (21 total)
- ✅ Science & Research: NASA, CERN, Max Planck, Bell Labs, ISRO (22 total)
- ✅ Social Services: UNICEF, Bill & Melinda Gates Foundation, Red Cross (23 total)
- ✅ Hospitality & Tourism: Airbnb, Marriott, Expedia, Hilton (21 total)
- ✅ Business & Management: McKinsey, BCG, Deloitte, Accenture (23 total)
- ✅ Sales & Marketing: Ogilvy, McCann, Leo Burnett, Dentsu (20 total)
- ✅ Agentic Intelligence: Multi-agent orchestration for real-time web research with Evergreen Perpetual Freshness.
- ✅ Stealth Startup Support: AI can "reverse-engineer" company DNA from any Job Description (JD)
- ✅ Synthetic Logic: Automatic industry-standard fallback if no web data is found
- ✅ Permanent Memory: Dynamic profile saving to discoveries.json
✅ Company-specific simulations (Google, Amazon, etc.)
✅ Adaptive difficulty levels (Junior, Mid-level, Senior)
✅ Panel interview mode with multiple AI personas
✅ Context-aware questioning based on resume and JD

📄 Resume Analysis (Backend Ready)

✅ ATS Score calculation (0-100)
✅ Gap analysis vs job descriptions
✅ Strengths & weaknesses identification
✅ Actionable improvement tips
⚠️ UI display pending

📊 Evaluation System

✅ No sugarcoating - Direct, professional feedback
✅ Behavioral vibe analysis (confidence, hesitation, assertiveness)
✅ Multi-dimensional scoring
✅ Comprehensive feedback after 5+ questions

🔐 Secure Authentication

✅ Secure JWT-based authentication system
✅ Protected API endpoints and user-specific sessions
✅ SHA-256 + Bcrypt double-hashing for maximum password security

🎨 Premium UI/UX

✅ Glassmorphism effects with advanced blur
✅ Animated mesh gradients
✅ 3D card transforms with magnetic hover
✅ Shimmering borders and gradient animations
✅ Professional meeting interface (Zoom/Meet inspired)
✅ Compact 64px toolbar
✅ Shimmer effects on primary buttons

🚀 Quick Start

Prerequisites

Python 3.11+
Node.js 18+
PostgreSQL 14+
Gemini API Key (Get one here)

Installation

1. Clone the Repository

git clone https://github.com/yourusername/interview-prep.git
cd interview-prep

2. Backend Setup

cd backend

# Create virtual environment
python -m venv venv

# Activate virtual environment
# Windows:
.\venv\Scripts\activate
# Mac/Linux:
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Create .env file
echo "GEMINI_API_KEY=your_api_key_here" > .env
echo "DATABASE_URL=postgresql://neondb_owner:[pass]@[endpoint].neon.tech/neondb?sslmode=require" >> .env

# Run database migrations (Syncs with Neon Cloud)
alembic upgrade head

# Start the backend server
python main.py

Backend will run on http://127.0.0.1:8000

3. Frontend Setup

cd frontend

# Install dependencies
npm install

# Start development server
npm run dev

Frontend will run on http://localhost:5173

🛠️ Tech Stack

Backend

FastAPI - Modern, fast web framework
SQLAlchemy - SQL toolkit and ORM
Alembic - Database migration tool
PostgreSQL - Robust relational database
PyPDF - PDF text extraction
Google Gemini 2.0 Flash - Advanced AI model

Frontend

React 18.3 - UI library
Vite - Next-generation frontend tooling
Axios - HTTP client
Vanilla CSS - Custom styling with glassmorphism
Web Speech API - Voice recognition

Planned Infrastructure

GitHub - Source control & CI/CD
Vercel - Frontend deployment
Neon - Serverless PostgreSQL Database (Cloud)
UPI QR Code - Direct payment (Zero fees)
Let's Encrypt - SSL certificates (Free)

📁 Project Structure

interview-prep/
├── backend/
│   ├── alembic/              # Database migrations
│   ├── services/
│   │   ├── gemini_service.py # AI service layer
│   │   └── company_intelligence.py # Company intelligence service (NEW)
│   ├── data/
│   │   └── company_profiles.json # Curated company database (NEW)
│   ├── models.py             # SQLAlchemy models
│   ├── schemas.py            # Pydantic schemas
│   ├── database.py           # Database configuration
│   ├── round_config.py       # Multi-round configuration
│   ├── main.py               # FastAPI application
│   ├── test_company_intel.py # Company intelligence test script (NEW)
│   └── requirements.txt      # Python dependencies
├── frontend/
│   ├── src/
│   │   ├── pages/
│   │   │   ├── Dashboard.jsx # Main dashboard component
│   │   │   ├── LandingPage.jsx # Landing page
│   │   │   └── Login.jsx     # Login/Signup
│   │   ├── App.css           # Setup screen styles
│   │   ├── Meeting.css       # Interview screen styles
│   │   └── InterviewerCards.css # Card component styles
│   ├── index.html            # HTML entry point
│   └── package.json          # Node dependencies
└── README.md                 # This file

🔌 API Documentation

Base URL

http://127.0.0.1:8000

Endpoints

`GET /health`

Health check endpoint

Response:

{
  "status": "ok",
  "version": "1.0.1"
}

`POST /interviews/start`

Start a new interview session

Request Body:

{
  "role_category": "Engineering & Tech",
  "sub_role": "Full Stack Developer",
  "difficulty_level": 1,
  "target_company": "Google",
  "is_panel": false,
  "interviewer_name": "Adinath"
}

Response:

{
  "id": 1,
  "first_question": "Good evening! I am Adinath, simulating a Technical interview..."
}

`POST /interviews/upload-resume`

Start interview with resume analysis

Form Data:

file: PDF file
role_category: string
sub_role: string
difficulty_level: integer (1-3)
target_company: string (optional)
interviewer_name: string

Response:

{
  "id": 1,
  "first_question": "...",
  "resume_analysis": {
    "ats_score": 85,
    "strengths": ["Strong technical background", "..."],
    "weaknesses": ["Missing cloud certifications", "..."],
    "tips": ["Add more quantifiable achievements", "..."]
  }
}

`POST /interviews/submit-answer`

Submit an answer and get next question or evaluation

Request Body:

{
  "interview_id": 1,
  "answer": "I have 3 years of experience in full-stack development..."
}

Response (Questions 1-4):

{
  "evaluation": null,
  "next_question": "That's interesting. Can you elaborate on...",
  "terminated": false,
  "questions_asked": 2
}

Response (After 5+ questions):

{
  "evaluation": {
    "score": 7.5,
    "feedback": "Strong technical knowledge but needs more confidence...",
    "vibe_analysis": {
      "confidence_score": 7,
      "hesitation_level": "Medium",
      "assertiveness": "Could be more decisive in answers"
    },
    "can_proceed": true
  },
  "next_question": null,
  "terminated": true,
  "questions_asked": 5
}

🎯 Usage Guide

Starting an Interview

Navigate to the setup screen
Choose your interviewer (Adinath or Veda)
Select your role category (12 categories available)
Pick a sub-role (specific to your category)
Set difficulty level (1=Junior, 2=Mid, 3=Senior)
Optional: Upload resume (PDF) for personalized questions
Optional: Add target company and job description
Optional: Enable panel mode for multi-interviewer experience
Click "ENTER MEETING ROOM"

During the Interview

Microphone: Auto-enabled for voice input
Camera: Disabled by default - AI will ask you to enable it
Voice Recognition: Speak naturally, AI transcribes in real-time
Submit Response: Click when you've finished your answer
Leave: Exit the interview at any time

After the Interview (5+ questions)

View your comprehensive evaluation
See your score out of 10
Read detailed feedback on your performance
Get behavioral analysis of your communication style
Review the complete transcript (future feature)

🔐 Environment Variables

Create a .env file in the backend directory:

# Required (ONLY PAID SERVICE)
GEMINI_API_KEY=your_gemini_api_key_here

# Database - Development (Local PostgreSQL)
DATABASE_URL=postgresql://user:password@localhost:5432/interviewai

# Database - Production (Supabase - will switch during deployment)
# DATABASE_URL=postgresql://postgres:[password]@db.[project-ref].supabase.co:5432/postgres

# Optional (for future features)
SECRET_KEY=your_secret_key_for_jwt
DEBUG=True

# UPI Payment (FREE - Your UPI ID)
UPI_ID=yourname@paytm
UPI_NAME=Your Name

Note: Currently using local PostgreSQL (pgAdmin4). Will migrate to Supabase during deployment phase.

💰 Cost Breakdown

Development Phase (Current)

Gemini API: ~₹0.10 per interview
Backend: ₹0 (Local development)
Database: ₹0 (PostgreSQL via pgAdmin4)
Frontend: ₹0 (Local development)
Total: ~₹50-100/month for Gemini API

Production Phase (After Deployment)

Gemini API: ~₹0.10 per interview (~₹100-500/month)
AWS EC2: ₹0 (Free tier 12 months), then ~₹500/month (t2.micro)
Supabase DB: ₹0 (Free tier - 500MB, 2GB bandwidth)
Vercel: ₹0 (Free tier)
SSL Certificates: ₹0 (Let's Encrypt)
GitHub: ₹0 (Free for public repos)
Payment Processing: ₹0 (Direct UPI)
Total Year 1: ~₹100-500/month (Gemini only)
Total Year 2+: ~₹600-1000/month (Gemini + AWS)

Revenue Model (After Payment Integration)

Free Tier: ₹0 (1 interview/2 weeks)
Pro Tier: ₹199/month (Unlimited interviews)
Elite Tier: ₹499/month (All features)
Break-even: ~3-5 Pro users or 2 Elite users per month

� Current Limitations & Known Issues

Limitations

⚠️ Single round only (Technical/Screening)
⚠️ No interview history tracking
⚠️ Resume analysis not displayed in UI
⚠️ No deployment (local only)
⚠️ No payment integration
⚠️ No learning roadmap generation

Known Issues

None currently reported ✅

📚 Project Documentation

Explore the deeper technical reasoning and project history:

📜 Activity Log & Collaborators - Track all major implementations and fixes.
🧠 Technical Challenges & Solutions - Deep dive into how we solved complex Agentic problems.
🗺️ Future Roadmap - See where the platform is headed next.

🤝 Contributing

Contributions are welcome! Please follow these steps:

Fork the repository
Create a feature branch (git checkout -b feature/AmazingFeature)
Commit your changes (git commit -m 'Add some AmazingFeature')
Push to the branch (git push origin feature/AmazingFeature)
Open a Pull Request

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

Google Gemini Team for the powerful AI model
FastAPI for the excellent web framework
React Team for the robust UI library
Inspired by industry-leading platforms like Zoom, Google Meet, and Linear

📧 Contact

For questions or support, please open an issue on GitHub.

Built with ❤️ for those who want to be more than just 'prepared'

🚀 Journey Status: Phase 1 Complete - Massive Challenges Ahead!

We've cleared the first hurdle. More logic, more code, more challenges await. 💪

⭐ Star this repo if you found it helpful!

Name		Name	Last commit message	Last commit date
Latest commit History 188 Commits
backend		backend
docs		docs
frontend		frontend
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md

Folders and files

Latest commit

History

Repository files navigation

🎯 InterviewAI - Professional Interview Intelligence Platform

🌟 The Simulation Philosophy

🎭 The Persona Architecture

🚀 Technical Innovation: The "Agentic Edge" (v2.3 — Domain-Aware Round Engine)

🛡️ Confidence Score Guide

🏗️ System Architecture

✅ Development Progress

Phase 1: Core Foundation (COMPLETED ✅)

Backend Infrastructure

Frontend Foundation

AI Interview System

Premium UI/UX

🚧 Roadmap - What's Next

🔥 Phase 2: Multi-Round Interview System (PENDING 🔴)

Round 1: Technical/Screening Round

Round 2: Behavioral/HR Round

Round 3: System Design Round (Technical Roles)

Round 4: Managerial Round

Round 5: Final/Director Round

Non-Technical Interview Rounds

� Phase 3: Advanced Analytics & Scoring (PENDING 🔴)

ATS & Resume Analysis

Performance Analytics

Behavioral Analysis

🎓 Phase 4: Learning & Growth (PENDING 🔴)

Post-Interview Learning

Skill Development

🔐 Phase 5: User Management & Authentication (IN PROGRESS 🟡)

User System

Pricing Tiers & Payment

UPI Payment Flow (Zero Cost)

☁️ Phase 6: Deployment & Infrastructure (PENDING 🔴)

Backend Deployment (AWS)

Database Migration

Frontend Deployment

DevOps Pipeline

🚀 Phase 7: Advanced Features (FUTURE 🔵)

AI Enhancements

Collaboration Features

Integration & API

✨ Current Features (v1.0.1)

🎤 Live Interview Simulation

🧠 Intelligent Interview System

📄 Resume Analysis (Backend Ready)

📊 Evaluation System

🔐 Secure Authentication

🎨 Premium UI/UX

🚀 Quick Start

Prerequisites

Installation

1. Clone the Repository

2. Backend Setup

3. Frontend Setup

🛠️ Tech Stack

Backend

Frontend

Planned Infrastructure

📁 Project Structure

🔌 API Documentation

Base URL

Endpoints

GET /health

POST /interviews/start

POST /interviews/upload-resume

POST /interviews/submit-answer

🎯 Usage Guide

Starting an Interview

During the Interview

After the Interview (5+ questions)

🔐 Environment Variables

💰 Cost Breakdown

Development Phase (Current)

Production Phase (After Deployment)

Revenue Model (After Payment Integration)

� Current Limitations & Known Issues

Limitations

`GET /health`

`POST /interviews/start`

`POST /interviews/upload-resume`

`POST /interviews/submit-answer`

Packages