AI Orchestrator

A production-quality AI Platform for executing multiple AI tasks through a single API. Built with modern Python practices, Clean Architecture principles, and provider-agnostic design.

Overview This platform provides a unified API for various AI tasks including SEO generation, documentation generation, text summarization, and more. The architecture is designed to be provider-agnostic, allowing seamless switching between different AI providers (Baseten, OpenAI, Anthropic, etc.) without modifying the core application.

Key Features

• ▸Multi-Task Support: SEO optimization, code documentation, text summarization, and extensible task system
• ▸Provider Agnostic: Easy switching between AI providers (Baseten, OpenAI, Anthropic, etc.)
• ▸Clean Architecture: Separation of concerns with dedicated layers for validation, execution, parsing
• ▸Production Ready: Comprehensive error handling, structured logging, retry logic, and monitoring
• ▸Modern Tech Stack: Python 3.12+, FastAPI, Pydantic v2, Docker, TypeScript frontend
• ▸Type Safety: Full type hints with mypy validation
• ▸Testing: Comprehensive test suite with pytest
• ▸Docker Support: Containerized deployment for easy scaling

Architecture The platform follows Clean Architecture principles with clear separation of concerns:

Request → Validation → Task Registry → Prompt Builder → Provider Layer → Response Parser → Response

Core Components

• ▸API Layer: FastAPI endpoints with request validation
• ▸Task Executor: Orchestrates task execution with error handling and retry logic
• ▸Task Registry: Dynamic task discovery and registration
• ▸Prompt Builders: Flexible prompt generation with template support
• ▸Provider Layer: Abstraction for different AI providers
• ▸Response Parsers: Structured response parsing and validation

Technology Stack Backend:

• ▸Python 3.12+: Modern Python with latest features
• ▸FastAPI: High-performance async API framework
• ▸Pydantic v2: Data validation and settings management
• ▸httpx: Async HTTP client for provider APIs
• ▸uv: Fast Python package manager
• ▸Docker: Containerization for deployment
• ▸Ruff: Fast Python linter
• ▸mypy: Static type checking
• ▸python-dotenv: Environment variable management
• ▸Structured logging: JSON-formatted logs for production monitoring

Frontend:

• ▸Next.js 14: React framework with App Router
• ▸TypeScript: Type-safe frontend development
• ▸Tailwind CSS: Utility-first CSS framework
• ▸React Hooks: Modern state management

Installation Prerequisites: Python 3.12+, uv package manager, Docker (optional), Node.js 18+ (for frontend)

Backend Setup:

git clone https://github.com/Othmane-aoubid/ai_orchestrator.git
cd ai_orchestrator
uv sync
cp .env.example .env

Running the Application Development Mode:

uv run uvicorn src.api.main:app --reload --host 0.0.0.0 --port 8000

Production Mode:

uvicorn src.api.main:app --host 0.0.0.0 --port 8000

Docker:

docker build -t ai-orchestrator .
docker run -p 8000:8000 --env-file .env ai-orchestrator

Testing

uv run pytest
uv run pytest --cov=src --cov-report=html

API Usage Execute a Task:

curl -X POST http://localhost:8000/api/v1/execute \
  -H "Content-Type: application/json" \
  -d '{
    "task_id": "seo",
    "inputs": {
      "content": "Your content here",
      "target_audience": "general audience"
    }
  }'

Available Tasks

• ▸seo: SEO content optimization
• ▸documentation: Code documentation generation
• ▸summarization: Text summarization with key points

License MIT License