AI Interviewer System

An intelligent, AI-powered interview platform that provides realistic interview experiences using advanced AI models and real-time conversational interfaces.

🚀 Features

Core Capabilities

• AI-Powered Interviews: Fully automated interviews powered by multiple AI providers
• Real-time Conversations: Live chat-like interface between AI and candidates
• Adaptive Questioning: AI adjusts questions based on candidate responses
• Persistent Interview State: Pause and resume interviews seamlessly
• Professional Audio Responses: AI-generated speech using ElevenLabs TTS
• Multi-language Support: English and Indonesian language support

Interview Management

• Customizable Interview Setup:
  • Candidate name and profile
  • Job role selection (e.g., backend-developer)
  • Difficulty levels (beginner, intermediate, expert)
  • Number of questions configuration
  • Company context for personalized interviews
  • Interviewer tone adjustment (formal, casual, etc.)

AI Integration

• Multiple AI Providers Support:

  • Z AI (GLM-4.6) - Primary provider
  • OpenAI (GPT models)
  • Anthropic (Claude models)
  • Mistral AI
  • Groq
  • XAI (Grok)
  • Google Gemini
  • DeepSeek
  • Ollama (local models)
  • OpenRouter
  • And many more

• Intelligent Conversation Flow:

  • Contextual understanding throughout interviews
  • Dynamic decision making for next actions
  • Follow-up question generation
  • Clarification requests
  • Natural, human-like conversations

User Experience

• Real-time Updates: Live message updates without page refresh
• Audio Playback: Automatic speech synthesis for AI responses
• Responsive Design: Works seamlessly on desktop and mobile
• Professional Interface: Clean, intuitive chat-like interface
• Loading States: Visual feedback during AI processing

🏗️ Architecture

Technology Stack

• Backend: Laravel 11.x with PHP 8.0+
• Frontend: Livewire + Volt + Tailwind CSS
• Build Tool: Vite
• Database: SQLite/MySQL/PostgreSQL
• AI Framework: Prism (multi-provider AI management)
• Testing: Pest
• Architecture: Domain-Driven Design (DDD)

System Architecture

├── Domain Layer (AI/Interviewer/Domain/)
│   ├── Primitives (Value Objects)
│   ├── Contracts (Interfaces)
│   └── Entities
├── Application Layer (AI/Interviewer/Application/)
│   ├── Data (DTOs)
│   └── Use Cases
└── Infrastructure Layer (AI/Interviewer/Infrastructure/)
    ├── Laravel Integration
    ├── Services
    └── Schemas

Database Schema

• Users: Authentication and user management
• Applicants: Interview candidate data with UUID primary keys
• ApplicantMessages: Conversation history with metadata

📦 Installation

Prerequisites

• PHP 8.0 or higher
• Composer
• Node.js and NPM
• SQLite/MySQL/PostgreSQL

Setup Steps

1. Clone the repository

   git clone <repository-url>
   cd ai-interviewer

2. Install dependencies

   composer install
   npm install

3. Environment setup

   cp .env.example .env
   php artisan key:generate

4. Configure environment variables

   # Database
   DB_CONNECTION=sqlite
   DB_DATABASE=database/database.sqlite
   
   # File Storage
   FILESYSTEM_DISK=public
   
   # AI Providers (at least one required)
   Z_API_KEY=your_z_api_key
   OPENAI_API_KEY=your_openai_key
   ANTHROPIC_API_KEY=your_anthropic_key
   ELEVENLABS_API_KEY=your_elevenlabs_key
   
   # Optional: Add other provider keys
   MISTRAL_API_KEY=your_mistral_key
   GROQ_API_KEY=your_groq_key
   GEMINI_API_KEY=your_gemini_key

5. Database setup

   php artisan migrate

6. Build assets

   npm run build

7. Link storage

   php artisan storage:link

8. Start the development server

   php artisan serve

🔧 Configuration

AI Provider Configuration

The system supports multiple AI providers through the Prism framework. Configure your preferred providers in config/prism.php:

'providers' => [
    'z' => [
        'api_key' => env('Z_API_KEY'),
        'base_url' => 'https://api.z.ai/api',
    ],
    'openai' => [
        'api_key' => env('OPENAI_API_KEY'),
    ],
    // Add other providers as needed
],

Default AI Provider

Set your default AI provider in the InterviewerService:

// src/Interviewer/Infrastructure/Services/InterviewerService.php
private string $provider = 'z';  // Change to your preferred provider
private string $model = 'glm-4.6'; // Change to your preferred model

TTS Configuration

For audio responses, configure ElevenLabs in your .env:

ELEVENLABS_API_KEY=your_elevenlabs_key

📖 Usage

Creating an Interview

1. Register/Login to the application
2. Navigate to the Interview Dashboard
3. Click "Start New Interview"
4. Fill in the interview configuration:
   • Candidate name
   • Job role
   • Difficulty level
   • Number of questions
   • Company context
   • Interviewer tone
5. Click "Generate Interview Link"

Interview Process

1. Welcome Phase: AI introduces itself and sets expectations
2. Question Phase: AI asks structured interview questions
3. Response Phase: Candidate provides responses
4. Analysis Phase: AI analyzes responses and determines next action:
   • Move to next question
   • Ask follow-up questions
   • Request clarification
   • Conclude interview

Features During Interview

• Real-time Messages: Instant display of AI responses
• Audio Playback: AI responses are automatically spoken
• Progress Tracking: Current question and completion status
• Persistent State: Interview can be interrupted and resumed

🧪 Testing

Run the test suite:

# Run all tests
php artisan test

# Run specific test file
php artisan test tests/Feature/AuthenticationTest.php

# Run with coverage
php artisan test --coverage

🔍 Monitoring & Debugging

Viewing Interview History

All interviews and conversations are stored in the database:

# View all applicants
php artisan tinker
>>> App\Models\Applicant::all();

# View interview messages
>>> App\Models\Applicant::find('uuid')->messages;

Debugging AI Responses

Structured responses include metadata for debugging:

{
  "message": "AI response text",
  "action": "ask_question|ask_followup|complete_interview",
  "status": "waiting_for_answer|question_asked|completed",
  "is_question": true,
  "move_to_next_question": true,
  "meta": {
    "provider": "z",
    "model": "glm-4.6"
  }
}

🚀 Deployment

Production Setup

1. Environment Configuration:

   APP_ENV=production
   APP_DEBUG=false

2. Optimize Application:

   php artisan config:cache
   php artisan route:cache
   php artisan view:cache
   php artisan optimize

3. File Permissions:

   chmod -R 755 storage
   chmod -R 755 bootstrap/cache

4. Queue Configuration (optional):

   php artisan queue:work

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

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

🙏 Acknowledgments

• Prism Framework: For providing multi-AI provider support
• Livewire: For enabling reactive UI components
• ElevenLabs: For high-quality text-to-speech synthesis
• Laravel Community: For the excellent framework and ecosystem

📞 Support

For support and questions:

• Create an issue in the GitHub repository
• Check the documentation
• Review the existing issues for common solutions

---

Built with ❤️ using Laravel and modern AI technologies.