Technology, AIIn the Era of ChatGPT and friendsIf you can no longer tell whether ChatGPT or a human wrote a piece of text, what does that do to authenticity and trust in what we read?
Technology, AIBing's great comebackMicrosoft wired ChatGPT into Bing and used the waitlist to push Bing onto millions of devices — pure marketing genius in my book.
Technology, AI, OpinionsDeepSeek R1 and the week the pricing story brokeWhen DeepSeek dropped R1, my first reaction was not hype — it was spreadsheet panic. The comfortable pricing story suddenly had a counterexample.
Technology, AI, CodeThe week coding assistants stopped pretending to be autocompleteClaude 3.7 Sonnet was the first time I caught myself delegating context, not lines. AI coding assistants stopped pretending to be autocomplete.
AI, Opinions, CodeAI code review is the bottleneck — writing got cheapAgents fill the merge queue faster than humans can read. The constraint moved from typing to verification — and most teams still review like nothing changed.
Technology, AI, Code, OpinionsVibe Coding: An Honest Take After Three Years of AI-Assisted DevelopmentThree years of AI-assisted development later, vibe coding is real. The thinking is yours, the judgment is yours — the typing was never the point.
AI, Code, ToolsUsing Claude Code's parallel agents to collaborate on a featureThe single-agent model has a ceiling. Claude Code's parallel agents let you split a feature into independent pieces and work like a focused team.
Ionic, Technology, AI, Code, PersonalThree similar Ionic apps, one Nx monorepo, and a lot of AI-assisted glue workHow we merged three Ionic/Angular/Capacitor codebases into a single Nx workspace so a small team could stop maintaining the same code in three places.
AI, Code, Testing, OpinionsGuardrails beat guidelines — how to keep AI-generated code honestYou cannot prompt AI-generated code into being reliable. What works is what always worked with new hires — guardrails that say no.
AI, Code, ToolsLet the big model write the manual — using Fable 5 to make smaller models betterI stopped pointing the frontier model at my backlog and started using it to write the Skills and guidelines my cheaper models read.
AI, Tools, TipsOne AGENTS.md for every coding agent — stop maintaining three rule filesPut shared project rules in AGENTS.md, keep a thin CLAUDE.md that imports it, and leave Skills for verbs — so Cursor, Claude Code, and Codex stop drifting apart.
AI, Tools, TipsBuilding a second brain in Obsidian — and telling my AI agents what not to write thereI finally built an Obsidian second brain that stuck — by writing down what my AI agents are and aren't allowed to put in it.
AI, Tools, Code, OpinionsValidation skills — the part of harness engineering that let me stop babysitting agentsHarness engineering gets agents doing the work. Validation skills are what let me trust that work enough to stop watching every run.
AI, Tools, Code, TechnologyMy Canon G5040 had no Mac app, so Claude Code reverse-engineered oneCanon ships no Mac utility for the Canon G5040. In one session, Claude Code reverse-engineered its maintenance protocol and built me one.
Opinions, AI, Code, TechnologySoftware is dead, long live the software — notes from 30 personal appsEveryone says software is dead now that anyone can build their own apps. I've built over 30 of them, and that's why I don't believe it.
