The Python workflow I use to catch bugs before production
How I set up Python for data migrations and site audits: Black and Ruff for formatting and linting, Mypy for data shapes, pre-commit hooks as a gate, and py-spy when something runs slow.
Category: Development
Recursive reasoning: how a 7M model beat DeepSeek R1
A 7M-parameter Tiny Recursion Model beat DeepSeek R1 on the ARC-AGI benchmark by looping over the problem instead of scaling up. What that says about trading parameters for iterations.
WordPress Studio adds a CLI and phpMyAdmin access
Automattic shipped two updates to WordPress Studio: a standalone CLI you install from npm, and phpMyAdmin access from the Overview tab. Here is how to turn both on and where each one fits next to the desktop app.
Linear regression as a projection in column space
Why the column space view of linear regression beats the scatter plot: features as vectors, the intercept as a vector of ones, and the normal equation as the point where the error runs perpendicular to the feature plane.
Running Claude Code agents in parallel with git worktrees
Most people run one Claude Code agent and then wait for it. Git worktrees let you keep several going in separate folders on the same repo. How I lay that out in Warp or Tmux, plus the shared local database problem nobody warns you about.
The CSS Olfactive API and its hardware problem
A read through the CSS Olfactive API draft: the scent-profile property, the whf potency unit, the prefers-reduced-pungency query, and why the missing hardware and browser support leave nothing to test yet.
Feature selection for credit scoring models in Python
How I pick features for credit scoring models in Python: Kruskal-Wallis for continuous variables, Cramer’s V for categorical ones, and Spearman rank correlation to catch redundant pairs before any training starts.
Building a custom sync provider in WordPress 7.0
WordPress 7.0 ships real-time collaboration over HTTP polling, which is fine until traffic picks up. Here is how I swap in a WebSocket transport with Yjs and the sync.providers filter, and what the auth side has to look like.
How I keep machine learning projects on track
Machine learning projects fail on planning more often than on algorithms. What works for me: chase approvals early, block real hours for the main pipeline, and stop repointing the stack every time a new model ships.
WordPress 7.0 APIs and why the release slipped
Notes from the Dev Chat on WordPress 7.0: the AI Client and Connectors API, why the release cycle got extended, the left navigation rethink, and the argument over YouTube support in the Cover block.