Setting Up Your AI Development Environment

Before you can build AI applications, you need a solid development environment. This lesson covers everything from Python setup and virtual environments to installing SDKs, managing API keys, and structuring your projects for success.

---

Why Environment Setup Matters

A well-configured development environment prevents the most common frustrations new AI developers face:

• Dependency conflicts between projects
• Leaked API keys that cost money or compromise security
• Inconsistent behaviour between your machine and production
• Wasted time debugging setup issues instead of building

---

Python Setup

Most AI libraries and SDKs have first-class Python support. We recommend Python 3.10+ for compatibility with modern AI tooling.

Checking Your Python Version

python3 --version
# Python 3.11.7  (or similar)

If you need to install or manage multiple Python versions, use pyenv:

# Install pyenv (macOS / Linux)
curl https://pyenv.run | bash

# Install a specific Python version
pyenv install 3.11.7
pyenv global 3.11.7

---

Virtual Environments

Always use a virtual environment for each project. This isolates dependencies and prevents conflicts.

Using venv (built-in)

# Create a virtual environment
python3 -m venv .venv

# Activate it
source .venv/bin/activate   # macOS / Linux
.venv\Scripts\activate      # Windows

# Verify
which python
# /path/to/project/.venv/bin/python

Using Poetry (alternative)

pip install poetry
poetry init
poetry add openai anthropic

---

Installing AI SDKs

The two most popular LLM providers have official Python SDKs:

Provider	SDK Package	Install Command
OpenAI	openai	pip install openai
Anthropic	anthropic	pip install anthropic

Quick Installation

pip install openai anthropic

Verifying Installation

import openai
import anthropic

print(f"OpenAI SDK version: {openai.__version__}")
print(f"Anthropic SDK version: {anthropic.__version__}")

---

API Keys

API keys are secrets that authenticate your requests. Never commit them to source control.

Getting Your API Keys

1. OpenAI: Go to platform.openai.com/api-keys and create a new key
2. Anthropic: Go to console.anthropic.com and generate a key

Environment Variables

Store keys in environment variables, not in your code:

# .env file (add to .gitignore!)
OPENAI_API_KEY=sk-proj-abc123...
ANTHROPIC_API_KEY=sk-ant-abc123...

Load them in Python using python-dotenv:

from dotenv import load_dotenv
import os

load_dotenv()  # reads .env file

openai_key = os.getenv("OPENAI_API_KEY")
anthropic_key = os.getenv("ANTHROPIC_API_KEY")

Security tip: Add .env to your .gitignore immediately after creating the file. If you accidentally commit a key, rotate it right away.

---

Project Structure

A clean project structure makes AI applications easier to maintain and extend:

my-ai-app/
├── .env                  # API keys (git-ignored)
├── .gitignore
├── requirements.txt      # or pyproject.toml
├── src/
│   ├── __init__.py
│   ├── client.py         # AI client setup
│   ├── prompts/
│   │   └── system.txt    # Prompt templates
│   ├── tools/
│   │   └── search.py     # Tool definitions
│   └── utils/
│       └── tokens.py     # Token counting, etc.
├── tests/
│   └── test_client.py
└── README.md

requirements.txt Example

openai>=1.0.0
anthropic>=0.18.0
python-dotenv>=1.0.0
tiktoken>=0.5.0

---

Your First API Call

With everything set up, let's verify it works:

from dotenv import load_dotenv
from openai import OpenAI

load_dotenv()
client = OpenAI()  # reads OPENAI_API_KEY automatically

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Say hello!"}
    ]
)

print(response.choices[0].message.content)

If you see a greeting in your terminal, your environment is ready.

---

Summary

• Use Python 3.10+ and manage versions with pyenv if needed.
• Always work inside a virtual environment (venv or Poetry).
• Install provider SDKs with pip install openai anthropic.
• Store API keys in environment variables via a .env file — never in code.
• Structure your project with separate directories for source, prompts, tools, and tests.
• Verify your setup with a simple API call before building anything complex.