GETTING_STARTED.md

FluxAI Gateway - Getting Started

Quick Start

Prerequisites

• Python 3.11+
• PostgreSQL 15+
• Redis 7+
• AWS Account with Bedrock access
• Docker and Docker Compose (optional, for containerized development)

Option 1: Docker Compose (Recommended for Development)

1. Clone the repository

   git clone https://github.com/AgentaFlow/fluxai.git
   cd fluxai

2. Configure environment

   cp .env.example .env
   # Edit .env and add your AWS credentials

3. Start services

   docker-compose up -d

4. Access the API

   • API: http://localhost:8080
   • Docs: http://localhost:8080/docs
   • Prometheus: http://localhost:9090
   • Grafana: http://localhost:3000 (admin/admin)

Option 2: Local Development

1. Clone the repository

   git clone https://github.com/AgentaFlow/fluxai.git
   cd fluxai

2. Create virtual environment

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Configure environment

   cp .env.example .env
   # Edit .env and configure:
   # - AWS credentials
   # - Database URL
   # - Redis URL

5. Start PostgreSQL and Redis

   # Using Docker:
   docker run -d -p 5432:5432 -e POSTGRES_PASSWORD=dev_password -e POSTGRES_DB=fluxai postgres:15
   docker run -d -p 6379:6379 redis:7-alpine

6. Run the application

   python -m app.main
   # Or with uvicorn directly:
   uvicorn app.main:app --reload --host 0.0.0.0 --port 8080

7. Access the API

   • API: http://localhost:8080
   • Interactive docs: http://localhost:8080/docs
   • Health check: http://localhost:8080/health

Testing the API

1. Health Check

curl http://localhost:8080/health

2. Invoke a Model

curl -X POST http://localhost:8080/v1/bedrock/invoke \
  -H "Content-Type: application/json" \
  -H "X-API-Key: your-api-key" \
  -d '{
    "model": "auto",
    "messages": [
      {
        "role": "user",
        "content": "Explain quantum computing in simple terms"
      }
    ],
    "max_tokens": 500,
    "temperature": 0.7
  }'

3. View Metrics

curl http://localhost:8080/metrics

Development

Project Structure

fluxai/
├── app/
│   ├── __init__.py
│   ├── main.py              # Application entry point
│   ├── api/                 # API routes
│   │   ├── v1/
│   │   │   ├── endpoints/
│   │   │   │   ├── bedrock.py
│   │   │   │   ├── analytics.py
│   │   │   │   ├── models.py
│   │   │   │   └── cache.py
│   │   └── dependencies.py
│   ├── core/                # Core configuration
│   │   ├── config.py
│   │   └── logging.py
│   ├── db/                  # Database
│   │   ├── models.py
│   │   └── session.py
│   ├── models/              # Pydantic schemas
│   │   └── schemas.py
│   └── services/            # Business logic
│       ├── bedrock_client.py
│       ├── cost_calculator.py
│       ├── cache.py
│       ├── metrics.py
│       └── analytics.py
├── requirements.txt
├── docker-compose.yml
├── Dockerfile
├── .env.example
└── README.md

Running Tests

pytest

Code Formatting

black app/
ruff check app/

Type Checking

mypy app/

Next Steps

1. Set up AWS credentials in .env
2. Review the Technical Specification for architecture details
3. Check the Implementation Guide for feature roadmap
4. Start building MVP features from the project plan

Need Help?

• Documentation: See markdown files in the project root
• API Docs: http://localhost:8080/docs (when running)
• Issues: GitHub Issues

License

See LICENSE file for details.