Kyle Kelley

@kylekelley.bsky.social

625 Followers 1062 Following

Quill 🎮 Anaconda 🐍 🧪 Jupyter & Zed Hacker ⌨️ Santa Cruz City Schools Board 🏫 Untitled.ipynb

nteract Settings window floating over a notebook. The settings panel shows Appearance (theme: Light/Dark/System), New Notebooks (default runtime: Python or Deno), Python Defaults (environment: uv or Conda, with packages matplotlib, numpy, requests), and Advanced settings with a Keep Alive slider set to 20 seconds.
Animated demo of Zed's REPL notebook output scroll containment. A Python traceback from a deeply recursive function overflows the output area, showing a "Click to scroll" overlay. The user clicks to enter scroll mode, scrolls through the full traceback with highlighted stack frames, then presses Escape to exit back to the editor. The output stays contained within the editor instead of pushing everything else off screen.
Custom keymaps in the zed editor. repl:run is triggered by keystrokes "x j", "x o", when Editor && VImControl is on.
Zed's inline REPL running a Lattice Boltzmann fluid simulation from a Python file. Code evaluates top-to-bottom — dicts and functions execute as single blocks with checkmarks appearing after each. A vorticity plot then animates live below the code, showing red and blue vortices shedding off a cylinder as the step counter climbs through 3000 iterations. Gruvbox Dark theme.
Screenshot of a desktop notebook app with a modal titled “Review Dependencies.” It lists PyPI packages pandas and numpy and asks the user to “Trust & Install” or “Don’t Install” before running code. The notebook UI behind it shows a dependency panel and Python cells importing those packages.
The user interacts with a slider that alters the frequency in the equation sin(freq*x), updating the plot using the new value. On the right is the python code that went to creating this setup. ``` In [1]: import runtimed In [2]: runtimed.sidecar() Out[2]: Sidecar(kernel-32461, running) In [3]: import numpy as np ...: import matplotlib.pyplot as plt ...: from ipywidgets import interact, FloatSlider ...: ...: x = np.linspace(0, 2 * np.pi, 200) ...: ...: @interact(freq=FloatSlider(min=0.5, max=5, step=0.1, value=1, description='Frequency')) ...: def plot(freq): ...: plt.figure(figsize=(8, 4)) ...: plt.plot(x, np.sin(freq * x), color='#6366f1', linewidth=2) ...: plt.ylim(-1.5, 1.5) ...: plt.title(f'sin({freq:.1f}x)') ...: plt.grid(True, alpha=0.3) ...: plt.show() ...: interactive(children=(FloatSlider(value=1.0, description='Frequency', max=5.0, min=0.5), Output()), _dom_class… ```
Split-screen view of a Jupyter environment. Left: Python code for setting up a sidecar viewer and loading data. Right: Data table displaying athlete information with 7 sample rows out of 11,538, showing fields like ID, name, nationality, and physical attributes.
Letter of acceptance for a workshop at the Annual Education Conference. "It is my pleasure to inform you that your Workshop session, ChatGPT as a Tool for Teaching and Learning, has been accepted to present at the 2023 Annual Education Conference and Trade Show in San Francisco. Title: ChatGPT as a Tool for Teaching and Learning Strand: The Promise and Peril of Technology Description: Santa Cruz City Schools worked with staff to discuss the value and possibilities that artificial intelligence (AI) and large language models like ChatGPT can provide to support teaching and learning. Staff are committed to teaching students how to responsibly use AI in learning. The district‘s teacher technology leaders are also committed to sharing how to utilize AI to enhance instruction and create efficiencies for teachers. In this workshop, hear from a trustee, teacher leaders and administrators about their experiences using AI in schools."