HireAbility - AI-Powered Mock Interview Platform

Transform your interviewing journey with AI-powered mock interviews

📋 Overview

HireAbility is an intelligent mock interview platform that leverages AI to provide realistic, personalized interview practice sessions. The platform conducts live AI-led interviews using audio and video, evaluates communication skills, technical accuracy, and confidence levels.

Key Features

• 🎤 Real-time AI Interviews Live voice and video interviews powered by VAPI, dynamically asking resume-based, role-specific, and company-specific questions.

• 🤖 Intelligent Evaluation Google Gemini analyzes your responses, voice patterns (clarity, pace, confidence), and facial expressions to assess communication and professional demeanor.

• 📊 Comprehensive Analytics Parameter-wise scores, strengths, weaknesses, and targeted improvement insights.

• 📄 Resume-Aware System Upload your resume for tailored questioning and AI-driven resume optimization.

• 🎯 Personalized Experience Fully customized interview flow based on your chosen role, company, and experience level.

🏗️ Architecture

Frontend (Next.js) → REST API → Backend (Express.js) → AI Agents (Gemini) → PostgreSQL
                    ↓
                 VAPI (WebSocket)

Tech Stack

Frontend: Next.js 16, React 19, TypeScript, Tailwind CSS, VAPI Web SDK
Backend: Express.js, TypeScript, Prisma ORM, PostgreSQL, JWT
AI: Google Gemini 1.5 Pro/Flash, VAPI Voice API

🚀 Setup

Prerequisites

• Node.js 18+, PostgreSQL 14+
• Google Gemini API Key
• VAPI API Key and Assistant ID

Environment Variables

Backend (be_hireability/.env):

DATABASE_URL="postgresql://user:password@localhost:5432/hireability"
JWT_SECRET="your-secret-key"
GEMINI_API_KEY="your-gemini-api-key"
MODEL_NAME="gemini-1.5-pro"
PORT=3001

Frontend (fe_hireability/.env.local):

NEXT_PUBLIC_API_URL=http://localhost:3001
NEXT_PUBLIC_VAPI_ASSISTANT_ID=your-vapi-assistant-id

Installation

# Clone repository
git clone <repository-url>
cd hireability

# Setup Backend
cd be_hireability
npm install
npx prisma generate
npx prisma migrate dev
npm run dev

# Setup Frontend (new terminal)
cd fe_hireability
npm install
npm run dev

Access: Frontend: http://localhost:3000 | Backend: http://localhost:3001

📡 API Endpoints

Base URL: http://localhost:3001/api

Authentication

• POST /api/auth/register - Register user
• POST /api/auth/login - Login user
• GET /api/auth/me - Get current user

Interviews

• POST /api/interviews - Start interview
• GET /api/interviews - Get all interviews
• GET /api/interviews/:id - Get interview details
• POST /api/interviews/:id/analyze - Analyze transcript
• GET /api/interviews/stats - Get statistics

Other

• POST /api/transcripts - Save transcript
• POST /api/documents - Upload resume
• GET /api/profile - Get/Update profile

Authentication: Include Authorization: Bearer <token> header

Example Request

POST /api/interviews
Authorization: Bearer <token>
Content-Type: application/json

{
  "assistantId": "vapi-assistant-id",
  "contextPrompt": "Interview for Software Engineer at Google"
}

🧪 How It Works

1. User starts interview → System creates session
2. VAPI connects → Real-time voice conversation
3. Transcript saved → Conversation stored
4. AI Analysis → Gemini analyzes across 6 dimensions:
   • Technical Skills (20%)
   • Problem Solving (20%)
   • Role Knowledge (20%)
   • Experience (15%)
   • Communication (15%)
   • Professional Demeanor (10%)
5. Feedback Generated → Scores, strengths, weaknesses, improvements

Sample Output

{
  "technical": {
    "score": 8.5,
    "strengths": ["Strong system design knowledge"],
    "weaknesses": ["Could improve trade-off discussions"],
    "improvements": ["Practice explaining trade-offs"]
  },
  "overall": {
    "score": 7.8,
    "summary": "Solid performance with good technical skills"
  }
}

📁 Project Structure

hireability/
├── be_hireability/     # Backend (Express.js)
│   ├── src/
│   │   ├── agents/     # AI agents
│   │   ├── controllers/
│   │   ├── services/
│   │   └── routes/
│   └── prisma/         # Database schema
│
└── fe_hireability/     # Frontend (Next.js)
    ├── app/            # Pages
    ├── components/     # React components
    └── lib/            # Utilities

📦 Key Dependencies

Backend: express, @prisma/client, @google/genai, jsonwebtoken, bcryptjs
Frontend: next, react, @vapi-ai/web, face-api.js, framer-motion

See package.json files for complete lists.

🚢 Deployment

Backend:

npm run build
npx prisma migrate deploy
npm start

Frontend:

npm run build
npm start

Recommended: Backend (Render/Railway), Frontend (Vercel/Netlify), Database (Supabase/Neon)

🎯 Status

Completed (60-80%):

• ✅ Core AI interview analysis
• ✅ Real-time VAPI interviews
• ✅ Transcript processing
• ✅ Multi-dimensional feedback
• ✅ User auth & profiles
• ✅ Resume review

In Progress:

• 🔄 Enhanced analytics
• 🔄 Real-time feedback

🤝 Contributors

• shaunakc11 - Shaunak Choudhury
• 0xPixelNinja - Rakesh Kumar Rabhi
• gaghackz - Karanam Gagan Deep
• pranjal-kumar-0 - Pranjal Kumar

📄 License

MIT License

---

Built for IIT Bombay Hackathon 2025