Introductory AI learning materials for freshmen and community college students. All notebooks can run on CPU on a shared JupyterHub but run faster on GPU enabled Juptyerhubs.
This work was presented as part of a tutorial at the 2026 NAIRR Annual Meeting.
Resources
-
Slide Deck 1 - Intro to Team, Intro to Jupyterhub - Intro to Small Models
-
Slide Deck 2 - Teaching workflow, NRP, Jetstream2
1. **Start the kernel:** Click the **power button** (top right of the page). Wait until the kernel or environment is ready.
2. **Run code:** Use the **play button** next to a cell to run it, or the **Run all** play button (top right, follows you as you scroll) to run every cell on the page.
The first time you launch with Binder, building the environment can take a few minutes. Later visits are usually faster.
- **External link** (next to the play button): Opens the notebook in a **new browser tab** using [Binder](https://mybinder.org/) (free, no account required).
- **Rocket ship** (launch menu): Choose **Binder** or **Custom** and paste the URL of your own Jupyter Hub ([jupyter.cal-icor.org](https://jupyter.cal-icor.org), [datahub.berkeley.edu](https://datahub.berkeley.edu), etc.).
- **Download:** Use the **download** control to save the notebook (e.g. `.ipynb`).
- **View source:** Use the **GitHub icon** to open the source in the repository.