crazy4pi314

There's no place like home: using AI agents in Jupyter notebooks

PyData Seattle 2025

This workshop explores how AI agents integrated directly into Jupyter notebooks can help with every part of your data science work. We'll cover the latest notebook-focused agentic features in VS Code, demonstrating how they automate tedious tasks like environment management or graph styling, enhance your "scratch notebook" to sharable code, and more generally streamline data science workflows directly in notebooks.

crazy4pi314

1. On the page with the embedded calendar you want the ical feed for, you use browser tools to inspect the page. Find the iframe element for the calenar and look for the src property:

2. Copy that src url and then use the tip [here](https://stackoverflow.com/questions/30766477/google-calendar-ical-url-for-public-calendar) to transform it:

the src address in an embed link - either email or calendar group account (randomhash@group.calendar.google.com) - is part of the ical url

https://calendar.google.com/calendar/embed?src=**info@example.com**

>to

crazy4pi314

{"name":"Notepad","settings":"{\"settings\":\"{\\r\\n    \\\"window.zoomLevel\\\": 3,\\r\\n    \\\"editor.inlineSuggest.enabled\\\": true,\\r\\n    \\\"github.copilot.enable\\\": {\\r\\n        \\\"*\\\": true,\\r\\n        \\\"yaml\\\": false,\\r\\n        \\\"plaintext\\\": false,\\r\\n        \\\"markdown\\\": true\\r\\n    },\\r\\n    \\\"workbench.colorTheme\\\": \\\"Witch Hazel\\\",\\r\\n    \\\"cSpell.userWords\\\": [\\r\\n        \\\"Codespace\\\",\\r\\n        \\\"Codespaces\\\",\\r\\n        \\\"conda\\\",\\r\\n        \\\"cuda\\\",\\r\\n        \\\"devcontainer\\\",\\r\\n        \\\"Ferrie\\\",\\r\\n        \\\"liveshare\\\",\\r\\n        \\\"matplotlib\\\",\\r\\n        \\\"Micromamba\\\",\\r\\n        \\\"Miniconda\\\",\\r\\n        \\\"Miniforge\\\",\\r\\n        \\\"navivage\\\",\\r\\n        \\\"nbconvert\\\",\\r\\n        \\\"pyenv\\\",\\r\\n        \\\"Pylance\\\",\\r\\n        \\\"screencast\\\",\\r\\n        \\\"screencasts\\\",\\r\\n        \\\"tetris\\\"\\r\\n    ],\\r\\n    \\\"workbenc

crazy4pi314

How to use:

1. Make sure you have your OpenAI API key set as an environment variable (OPENAI_API_KEY)
2. Set up the pre-commit message prep hook bu adding the other two files here in the ./git/hooksdirectory.

Test using it by staging some changes to commit and then running git commit in the command line. The file it pops up for you to edit now should have the suggested summary of the changes from the OpenAI api commented out like this:

# Here is a suggested commit message from code-davinci-002.
# ---

crazy4pi314

crazy4pi314

FROM mcr.microsoft.com/devcontainers/miniconda:latest

# Copy environment.yml (if found) to a temp location so we update the environment. Also
# copy "noop.txt" so the COPY instruction does not fail if no environment.yml exists.
RUN conda install -n base -c conda-forge mamba
COPY environment.yml* .devcontainer/noop.txt /tmp/conda-tmp/
RUN if [ -f "/tmp/conda-tmp/environment.yml" ]; then umask 0002 && /opt/conda/bin/mamba env update -n base -f /tmp/conda-tmp/environment.yml; fi \
    && rm -rf /tmp/conda-tmp

# [Optional] Uncomment this section to install additional OS packages.

crazy4pi314

1. Open the qat codespace, and if it errors from being out of memory, locally clone and use remote dev containers to create docker
2. Build qat tool with instructions here: https://qir-alliance.github.io/qat/UserGuide/BuildingLibrary/#:~:text=Build%20instructions%20for%20users

mkdir Debug
cd Debug
cmake ..
make qat

3. Try running the built executable with some test QIR in the test folders here: https://github.com/qir-alliance/qat/tree/main/qir/qir-tests

crazy4pi314

crazy4pi314

crazy4pi314

// Configure properties specific to VS Code.
"vscode": {
  // Set *default* container specific settings.json values on container create.
  "settings": {
    "editor.fontSize": 14,
    "editor.lineHeight": 22,
    "window.zoomLevel": 4,
    "editor.formatOnSave": false,
    "editor.lineNumbers": "off",
    "editor.quickSuggestionsDelay": 1000,