November 7, 2026 · 7 min read
ADRs for AI Coding Agents. Preserving the Reasons Behind Architecture Decisions
Why AI coding agents need the reasoning behind architectural choices, and how to keep ADRs with completed tasks.
Read more →Posts
October 31, 2026 · 8 min read
Local code review with AI agents, without opening a pull request
A developer’s search for a local browser-based code review workflow for Claude Code and Codex, comparing difit, Plannotator, and diffx.
Read more →October 24, 2026 · 10 min read
AGENTS.md for Codex and Claude Code. What to Include and What to Move into Skills
A practical look at AGENTS.md, scoped project instructions, and when to keep detailed workflows in Skills instead.
Read more →October 17, 2026 · 15 min read
some vs any in Swift: The Difference and When to Use Each
Understanding existential and opaque types through checkout, SDK design, SwiftUI, and Swift runtime internals.
Read more →October 10, 2026 · 12 min read
How to Clean Xcode DerivedData and Why It Grows to Tens of Gigabytes
Why Xcode DerivedData grows to tens of gigabytes on a Mac, how to inspect and clean it (Xcode Settings, Terminal, Xcode 27, free tools), and a small zsh script called `xcode-dd` for a one-command overview.
Read more →August 29, 2026 · 16 min read
Where Codex Tokens Actually Go: 67% Tool Outputs, 4.6% My Prompts
A personal experiment with Codex session logs: where tokens actually go, how I measured usage, and what I changed in the way I manage context.
Read more →August 22, 2026 · 10 min read
iOS Logging and Analytics: A Practical Guide
Logging and analytics aren't a checkbox. They do different jobs, and if you don't understand which, all that instrumentation quietly turns into noise you can't get anything out of.
Read more →