Archive

2026

July

• 06 A Notebook That Forgets on Purpose

May

• 10 Polish Without Priors

April

• 22 What the Agent Noticed
• 15 `gco`: Your Brain Shouldn't Track What Git Already Knows

January

• 17 Why Alignment Verification Might Be Fundamentally Broken

2025

December

• 30 The Layers of Abstraction Will Kill You
• 10 I Made a Lightweight Git Worktree Manager (Because I Couldn't Find One)

November

• 26 Let the System Remember: Multiple Git Accounts
• 03 Syntax, Struggle, and Expertise

October

• 22 Review-First Programming
• 16 Document extraction: four main approaches with a 1000x cost difference
• 06 Semantic layers are finally getting opinionated enough to be useful
• 06 Semantic Layer Solutions in Modern Data Architecture

September

• 25 The Yard, The Sparkly Hat, and The Doomsday Clock
• 16 The LinkedIn Sharing Paradox: Why Testing Social Media Integration is Harder Than It Should Be
• 14 Multi-Head Attention: Full Input Projection Not Slicing
• 02 The Null Equivalence Principle for JSON APIs

August

• 26 猫(Neko)
• 22 Flask's `g` - The "Global" That Isn't
• 15 AWS DMS for Simple CDC Pipelines

July

• 25 Reranking
• 16 System Prompt Testing Methodology
• 09 From Fabric User to Pattern Creator: Building Better AI Workflows
• 02 Model Autophagy Disorder: AI Will Eat Itself

June

• 27 When the Robots Came for the Coders
• 20 against perfection, against overthinking, against the tyranny of having everything figured out before you begin
• 20 Simon Willison’s Weblog