Python Environment Setup for AI Engineering

Python Environment Setup for AI Engineering

A professional AI automation environment is not just "install Python and pip install." You need reproducible environments, type checking, async support, and structured project layouts that scale.

Step 1: Python Version Management with pyenv

# Install pyenv (macOS)
brew install pyenv

# Install Python 3.12
pyenv install 3.12.3
pyenv global 3.12.3

# Verify
python --version
# Python 3.12.3

Step 2: Project Structure

Every project in this course follows this structure:

ai-automation-project/
├── src/
│   ├── __init__.py
│   ├── config.py          # Environment and settings
│   ├── client.py          # API client setup
│   ├── pipelines/         # Automation pipelines
│   │   ├── __init__.py
│   │   └── document.py
│   ├── models/            # Data models (Pydantic)
│   │   ├── __init__.py
│   │   └── schemas.py
│   └── utils/             # Shared utilities
│       ├── __init__.py
│       ├── retry.py
│       └── logging.py
├── tests/
│   └── test_pipelines.py
├── .env                   # API keys (never commit)
├── .env.example           # Template for env vars
├── pyproject.toml         # Project metadata and deps
├── Dockerfile
└── README.md

Step 3: Dependency Management with uv

We use uv — the fastest Python package manager (written in Rust):

# Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh

# Create project
mkdir ai-automation && cd ai-automation
uv init

# Add core dependencies
uv add anthropic pydantic python-dotenv httpx tenacity structlog

# Add development dependencies
uv add --dev pytest pytest-asyncio ruff mypy

Step 4: Configuration with Pydantic Settings

# src/config.py
from pydantic_settings import BaseSettings

class Settings(BaseSettings):
    anthropic_api_key: str
    default_model: str = "claude-sonnet-4-5"
    max_retries: int = 3
    log_level: str = "INFO"

    class Config:
        env_file = ".env"
        env_file_encoding = "utf-8"

settings = Settings()

Step 5: The .env File

# .env
ANTHROPIC_API_KEY=sk-ant-your-key-here
DEFAULT_MODEL=claude-sonnet-4-5
LOG_LEVEL=INFO

Step 6: Structured Logging

# src/utils/logging.py
import structlog

def setup_logging(level: str = "INFO"):
    structlog.configure(
        processors=[
            structlog.processors.TimeStamper(fmt="iso"),
            structlog.processors.add_log_level,
            structlog.processors.StackInfoRenderer(),
            structlog.dev.ConsoleRenderer(),
        ],
        wrapper_class=structlog.make_filtering_bound_logger(level),
    )
    return structlog.get_logger()

log = setup_logging()

Step 7: Verify Everything Works

# test_setup.py
import anthropic
from src.config import settings

client = anthropic.Anthropic(api_key=settings.anthropic_api_key)

response = client.messages.create(
    model=settings.default_model,
    max_tokens=100,
    messages=[{"role": "user", "content": "Say 'Setup complete!' if you can read this."}],
)

print(response.content[0].text)
# Setup complete!

Run it:

uv run python test_setup.py

If you see "Setup complete!" — your environment is ready. Every project in this course builds on this foundation.