Welcome to John Maeda s Blog: maeda.pm. Sharing a passion for product and data, grounded in technology / business / design curiosities with a dash of leadership reflections.
For most of my career I designed the cockpit. I made the menus and the tabs and the flows. I arranged the dashboards and the confirmation dialogs and the little blinking things that tell you something is happening. I was proud of this work. And the better I made it, the more considered the controls [ ]
https://audio.raphaelsalaja.com
Source: https://simonwillison.net/2025/Dec/31/the-year-in-llms/
I started this thought in this post over here. Applied directly to NeonJS
1) Install prerequisites A) Homebrew B) llama.cpp via Homebrew Get llama.cpp via brew: Verify: C) ollama via Homebrew There s an easy installer now instead of doing it the brew way. Verify: D) Hugging Face CLI (for downloads) Verify: (Optional) login if you want gated models later: 2) Create a local workspace 3) Download a tiny [ ]
I wanted to use this https://github.com/fatih/dotfiles/blob/main/claude/statusline-git.sh 2. In ~/.claude/statusline.sh drop the code from the repo into there 3. chmod +x ~/.claude/statusline.sh 4. echo { model :{ display_name : Claude Sonnet 4.5 }, workspace :{ current_dir : $(pwd)' }} ~/.claude/statusline.sh 5. run claude and you ll see the new status line bar
Govind s Out Of Distribution Thinking note and some new LLM learnings Manifold Learning: Bing Wen Brunton Mechanistic Interpretability: Jenny Vega Visualizing LLMs Layers of FunctionGemma Unsloth FunctionGemma Notebooks Last Runpod State My Runpod Console
https://liner.com/news/fighting-startups-liner https://surgehq.ai
List of things I would have fun doing before I kick the bucket as a note to myself and for folks who are curious: