Installing Python

A short guide using uv

Why we use uv for this course

uv is a new (and very fast) Python tool written in Rust. It:

  • Installs Python for you (no manual downloads).
  • Creates isolated virtual environments (safe sandboxes per project).
  • Installs and updates packages quickly.

WHAT is a virtual environment? Think of each project as its own coffee shop with its own supplies. One shop changing its menu does not affect the others. WHY it matters: You avoid random breakage when different projects need different versions of the same package.

Install uv

Everything below happens in a terminal, a window where you type commands. On macOS open the Terminal app (Spotlight: Cmd+Space, type Terminal). On Windows open PowerShell (Start menu, type PowerShell, pick the entry called Windows PowerShell). To run a command, type or paste its line and press Enter. Choose the instructions for your operating system.

macOS or Linux (Terminal)

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

If curl is missing:

wget -qO- https://astral.sh/uv/install.sh | sh

After installation: close and reopen your terminal (so your PATH updates).

Windows (PowerShell)

Open PowerShell and run:

powershell -ExecutionPolicy Bypass -c "irm https://astral.sh/uv/install.ps1 | iex"

If you see a script execution warning, you can alternatively first run:

Set-ExecutionPolicy -ExecutionPolicy Bypass -Scope Process

Then re-run the install line.

After installation: close and reopen PowerShell, so that it finds the new uv command.

Verify installation

Run (macOS / Linux / Windows):

uv --version

If you see a version number: great!

If you get “command not found” or “‘uv’ is not recognized”:

  1. Close and reopen the terminal (important).
  2. On Windows: make sure you used PowerShell (not Command Prompt).
  3. Still broken? Ask for help, no need of guessing the error.

Install Python

We want everyone on the same Python version for consistency. Thus, we’ll use Python 3.12 for the course this year.

Install (you only need to do this once):

uv python install 3.12

Check the installation:

uv run python --version

Expected output starts with:

Python 3.12.

Create your first project

Pick a folder where you keep course work. If you do not have one, make sure to create one! In the terminal, move into that folder: type cd and a space, drag the folder from Finder or Explorer onto the terminal window (that pastes the folder’s location), and press Enter. Then run these two lines, each one once:

uv init --no-package python-labs
cd python-labs

The first line creates a new folder named python-labs. Keep that name: from Session VIII the labs live in this folder. The second line moves you into that folder. In Session X you will instead run uv init --no-package inside a folder cloned from GitHub; same command, no name.

uv init --no-package creates:

  • main.py (starter script)
  • pyproject.toml (project + dependencies config)
  • .python-version (records the Python version we chose)
  • README.md, and .gitignore if git is on your machine (only when the folder is new; a folder cloned from GitHub already has both)
  • (A .venv folder will appear later once packages are added or synced.)

Why --no-package? Without it, recent versions of uv build an installable package layout (src/python_labs/) with no main.py. For scripts, which is all this course needs, the flat layout is simpler.

You do not need to edit any of these (except maybe README.md for your notes and main.py if you want to run something different).

Run the starter script

Inside the project folder:

uv run python main.py

You should see something like:

Hello from python-labs!

(If you want, you can open main.py and change the message, then re-run.)

What does that code mean?

def main():
    print("Hello from python-labs!")


if __name__ == "__main__":
    main()
  • def main(): defines a function (a reusable block of code).
  • print(...) shows text in the terminal.
  • The line if __name__ == "__main__": ensures this only auto-runs when the file is executed directly.

Don’t worry about this yet, we’ll gradually build up to it.

Adding packages

The labs from Session VIII need two packages. Inside python-labs, run:

uv add pandas matplotlib

Check that both arrived. This line prints ready:

uv run python -c "import pandas, matplotlib; print('ready')"

If you added the wrong one (example: numpi, a typo):

uv remove numpi

If your pyproject.toml changed (e.g. you pulled code from someone else):

uv sync

If something seems “off”, just close the terminal and reopen in the project folder. Fresh starts fix many early mistakes.

Updating uv

Occasionally:

uv self update

(If it ever errors, you can just reinstall using the same one-liner from earlier.)

Best practices for this course

  • One folder for the course keeps everything tidy.
  • Never install packages “globally” outside a project.
  • Keep a short personal log in each project’s README.md (What did I do? What still confuses me?).
  • Ask early for help, guessing usually takes much more time than asking.

You can always see available commands:

uv --help

Recap

You can now:

  1. Install uv.
  2. Create a project.
  3. Run a script.
  4. Add/remove/sync packages.

Next come Zed and Mistral Vibe: AI inside Zed.