Long-form writing on Drupal, PHP frameworks, AI tooling, automated testing and everything else in a working programmer's stack — shipped from production, not theory.
The panic says every European developer needs a compliance officer by Sunday. I read the regulation instead. Using Claude Code to write code? Nothing changes for you. Shipping an LLM feature? Smaller than you fear. The €35M headline? Not even the right number. An honest split of hype from what a solo dev must actually care about.
METR ran a proper randomised trial: experienced devs felt AI made them 20% faster; the stopwatch said 19% slower. I use Claude Code daily and would have bet the same way they did. What the study actually measured, the nuance that stops it being 'AI is useless', and the uncomfortable thing I caught myself doing on code I know cold.
My OVH renewal came in ~45% higher, and for once the host isn't being greedy: the world's affordable RAM is being fed to AI accelerators. Why DRAM contract prices nearly doubled, why 'just switch to Hetzner' doesn't help, why it won't ease before 2027 — and the actual tuning that made my 4GB box need less memory instead.
Gmail and Outlook now hard-bounce unauthenticated mail instead of junking it. Sending straight from a VPS is a reputation treadmill you lose. The fix is to relay but keep your own SPF/DKIM/DMARC — which is exactly where I found my own footgun: a live 'v=spf1 -all' with no relay include, my newsletter surviving on DKIM alone. Here's what each record does and how to not do what I did.
I shipped an llms.txt on this blog last week, then went looking for proof anything reads it. Google's John Mueller compares it to the old keywords meta tag; a 2026 study of ~38,000 AI-cited domains found zero citation advantage (p=0.85). Where it's genuinely useful, how to test it against your own logs, and why a helper file is a long shot on a web that already ignores robots.txt.
GitHub Copilot went token-metered on 1 June, Cursor split its usage pools, and Sonnet 5's intro API price expires 31 August. The flat monthly plan for AI coding is quietly over. Why it broke, what it does to a working dev's bill, and the one rule I now budget by.
Coolify shipped a first-party MCP server in v4.1.0, so I wired Claude Code to the VPS that runs this blog and deployed by asking. It works, which is the worrying part. What the tools actually do (no rollback), the thin confirmation model, and how I scope tokens so an agent can't restart prod off a misread sentence.