Jupyter Notebook
What is a Jupyter Notebook?
A Jupyter Notebook is an interactive coding environment that combines executable code blocks, markdown notes, and data visualizations all in a single file. You can write code, run it immediately to see results, take notes in markdown alongside your code, and visualize data—making it particularly useful for experimentation, iterative development, and data analysis.
The notebook format allows you to move fluidly between brainstorming ideas in markdown and writing executable code, running experiments and immediately seeing results without deploying to a production environment.
Why are Jupyter Notebooks popular in AI development?
Jupyter Notebooks are especially popular in data science and AI development because they support rapid experimentation. You can make a prompt change, run your evals, adjust a temperature setting, run your evals again, try a different model, and run evals once more—all within the same notebook session, potentially running dozens of experiments in a single day.
The notebooks also capture output data that can be committed to version control systems like GitHub. This creates a history of every experiment with its results, allowing teams to compare different approaches—similar to running A/B tests on prompts, models, or other AI parameters.
How do product teams use Jupyter Notebooks?
Jupyter Notebooks may serve as a collaboration tool where product managers and engineers work together on AI products. Engineers can write the code blocks while product managers experiment with prompts, models, and parameters, running tests independently to explore different options.
For teams new to notebooks, JupyterLab provides a better experience than VS Code's notebook interface. The tool is accessible to beginners—even those with limited coding experience can learn to use notebooks effectively, especially with guidance from AI assistants like ChatGPT or Claude.
Learn more:
- How I Designed & Implemented Evals for Product Talk's Interview Coach
- Behind the Scenes: Building the Product Talk Interview Coach
- Q&A Session: Building AI Evals for the Interview Coach
Related terms: