In plain terms
Jupyter Notebook lets you write Python code in your browser and run it cell by cell, seeing the output immediately. It's designed for data exploration, AI prototyping, and building workflows step-by-step — without needing to set up a full Python environment on your computer.
What you can do
- Write AI code and run it on the server's GPU
- Test LangChain pipelines interactively
- Process and analyse datasets
- Run fine-tuning jobs (QLoRA / Unsloth)
- Save your work as a reusable notebook file
- Mix code, results, charts, and explanations in one document
Open in your browser
Open Jupyter in your browser
Go to http://ai.local:8888. You'll see the file browser showing notebooks saved on the server.
Create a new notebook
Click New → Python 3 to create a new notebook. A blank notebook opens with one empty cell.
Write and run code
Click into a cell, type your Python code, then press Shift + Enter to run it. The output appears directly below the cell.
Try calling Ollama from Python
Paste this into a cell and run it:
import ollama
response = ollama.chat(model="llama3.1:8b", messages=[{"role": "user", "content": "Explain RAG in one sentence"}])
print(response["message"]["content"])
Save your work
Press Ctrl + S (or Cmd + S on Mac) to save your notebook. It's saved on the server and will be there next time you open Jupyter.