Blog Posts
Thoughts on software architecture, AI engineering, and lessons learned from building real-world systems.
An AI Wiped a Production Database in 9 Seconds. Don't Blame the AI.
A coding agent deleted a company's entire database — backups included — in nine seconds. Everyone called it “AI gone rogue.” The more uncomfortable truth: the AI did exactly what it was built to do. The real culprit was the blast radius.
AI Is Just a Hack. So Was Newton.
George Hotz says the whole AI route is a hack. He's right about the diagnosis — and wrong about the conclusion. Here's why a bounded, imperfect hack is exactly how progress works.
Loop Engineering Doesn't Remove the Human. It Promotes One.
Everyone read "loops mean no humans, agents running 24/7." That's the part they got wrong — and the honest catch nobody mentions about agents checking agents.
Most People Still Can't Use AI. That Gap Is Your Business.
Half the world has never touched AI. You use it daily. Two small client stories about why that asymmetry is the clearest business opportunity since the early internet.
Stop Paying Genius Rates for Junior Work: My 24/7 AI Team Across Four Providers
One Claude Max subscription used to run my whole agent team — until they went always-on. Here's the org chart I built instead, matching each LLM to the role it deserves.
Your OpenClaw + Claude Setup Just Broke. Here’s How to Keep Using Opus & Sonnet Without Extra Costs.
A complete guide to migrating from OpenClaw to Claude Code’s native tools — done in one day, using your existing Claude subscription.
WordPress → Next.js: Migrating a Renovation Company's Website
How I rebuilt Reno Stars from a sluggish WordPress site to a modern Next.js app — the real tradeoffs, challenges, and results.
I Kept Removing AI Until It Worked
How I built an invoice automation system for Reno Stars by learning — the hard way — that more AI agents just means more problems.