Setting Up Your Environment

The shape of your workspace

Before the commands, picture what you're building: an isolated project folder with its own Python environment, a secrets file the model reads from, a notebook or script to run, and a Git repo to save it all.

Step by step

1. Install Python 3.11+ Download it from python.org (or your package manager). Confirm it works:

   python --version
   # Python 3.11.x  (or newer)

2. Create a project and a virtual environment A virtual environment keeps this course's packages isolated from the rest of your system.

   mkdir sage && cd sage
   python -m venv .venv
   # activate it:
   # macOS / Linux:
   source .venv/bin/activate
   # Windows (PowerShell):
   .venv\Scripts\Activate.ps1

   When it's active, your prompt shows (.venv).
3. Install the packages We start with the OpenAI SDK and a helper to load secrets. (We'll add more per section.)

   pip install openai python-dotenv

4. Get an API key and store it safely Create a key in your provider's dashboard, then put it in a file named .env — never in your code:

   # .env  (this file stays on your machine only)
   OPENAI_API_KEY=sk-your-key-here

5. Make your first model call Create hello.py:

   import os
   from dotenv import load_dotenv
   from openai import OpenAI
   
   load_dotenv()                       # read .env
   client = OpenAI()                   # uses OPENAI_API_KEY automatically
   
   resp = client.chat.completions.create(
       model="gpt-4o",
       messages=[{"role": "user", "content": "In one sentence, say hello as Sage, a friendly study assistant."}],
   )
   print(resp.choices[0].message.content)

   Run it:

   python hello.py

   You should see a friendly one-liner. That's Sage's first breath. 🎉
6. Initialise Git — and protect your secrets

   git init
   printf ".venv/\n.env\n__pycache__/\n" > .gitignore
   git add .
   git commit -m "Section 1: hello, Sage"

   The .gitignore guarantees your key and environment never get committed.

Why the key goes in .env, not your code

It's tempting to paste your key straight into the script. Don't. Keys in source code get pushed to GitHub, copied into screenshots, and shared in chats — and a leaked key can run up real charges on your account. Keeping it in a git-ignored .env file means the secret lives only on your machine, while your code (which is safe to share) just reads it by name.

Scripts or notebooks?

Both work, and you'll use both. VS Code with the Python extension is great for building real files like hello.py. Jupyter notebooks are perfect for experimenting cell by cell — handy when you're poking at model output. Install whichever you prefer now; the course's code runs identically in either.