README

A Laravel/Filament dashboard for monitoring AI activities within Baudex, an AI platform for construction companies. The dashboard provides insight into model usage, token consumption, costs, and task executions.

📋 About the Project

What does this dashboard do?

Baudex is an AI platform that helps construction companies analyze project documents, risk analyses, and cost estimations. This dashboard monitors:

• Which AI models are used and how often
• Token usage and costs per model and per project
• Storage usage (vector stores and files)
• Performance of executed tasks

Tech Stack

• Backend: Laravel 12, PHP 8.2+, MySQL
• Frontend: Filament 4.0, Livewire, TailwindCSS 4
• API: OpenAI API (with Mock API for development)

📊 Database Structure

Models

Project

Contains Baudex projects for which AI analyses are performed.

• id, name, created_at, updated_at

Message

Stores all AI interactions including token information.

• id, project_id, type, metadata (JSON)
• input_tokens, input_cached_tokens, output_tokens
• created_at, updated_at

Task

Defines AI tasks such as "Risk Analysis" or "Cost Estimation".

• id, name, created_at, updated_at

TaskRun

Records each task execution with performance metrics.

• id, task_id, message_id, duration (seconds)
• created_at, updated_at

📁 Project Structure

app/
├── Filament/
│   ├── Pages/                 # Dashboard pages (AiDashboard, ModelsDetail, UsageDetail, JobsDetail)
│   └── Widgets/               # Dashboard widgets (Models, Usage, Storage, Jobs)
│
├── Models/                    # Eloquent models (Project, Message, Task, TaskRun)
│
├── Services/                  # Business logic
│   ├── AiModelService.php     # Model usage processing
│   ├── UsageService.php       # Token/cost calculations
│   ├── StorageService.php     # Storage statistics
│   └── FilterService.php      # Time-period filter logic
│
├── Repositories/              # Database queries
│   ├── MessageRepository.php
│   └── TaskRunRepository.php
│
├── OpenAI/
│   └── OpenAiClient.php       # OpenAI API wrapper
│
└── Jobs/                      # Asynchronous jobs
    ├── FetchVectorStoreJob.php
    └── FetchFileJob.php

config/
├── api.php                    # API endpoint configuration
└── pricing.php                # Model token pricing

database/
├── migrations/                # Database schema
├── factories/                 # Test data factories
└── seeders/                   # Database seeders

tests/
├── Unit/                      # Unit tests for individual classes
└── Integration/               # Integration tests for component interaction

🧪 Testing

The project uses PHPUnit for automated testing.

Test Coverage

• ✅ Services – Data transformation and aggregation logic
• ✅ Repositories – Database query correctness
• ✅ API Client – API response handling
• ✅ Widgets – Data processing for dashboard widgets
• ✅ Filters – Filter functionality

Running Tests

# Run all tests
php artisan test

🚀 Local Setup

Requirements

• PHP >= 8.2
• Composer >= 2.0
• Node.js >= 18.x
• MySQL >= 8.0

Installation Steps

# 1. Clone repository
git clone <repository-url>
cd portfolio

# 2. Install dependencies
composer install
npm install

# 3. Environment setup
cp .env.example .env
php artisan key:generate

# 4. Configure database in .env
# DB_CONNECTION=mysql
# DB_DATABASE=ai_monitoring
# DB_USERNAME=root
# DB_PASSWORD=your_password

# 5. Database setup
php artisan migrate
php artisan db:seed

# 6. Create admin user
php artisan make:filament-user

# 7. Build assets
npm run dev

# 8. Start development server
php artisan serve

# 9. Start API server
php artisan serve --port=3000

# 10. Queue worker for async jobs
php artisan queue:work --timeout=300 --sleep=3 --tries=3

Then log in using your credentials at: http://localhost:8000/admin

Development URLs

• Dashboard: http://localhost:8000/admin
• Login with the credentials created in step 6

📝 Dashboard Features

Widgets (Overview Page)

• Models: Top 3 most-used AI models
• Usage: Total token consumption and costs
• Storage: Vector store and file statistics
• Jobs: Task execution frequency

Detail Pages

• Models Detail: Requests and token usage per model over time
• Usage Detail: Usage per project and per model
• Jobs Detail: Executions, token usage, and response time per task

All pages support time-period filters and toggling between tokens and euros.