This is short on purpose.
What we collect
Feedback on AI answers. When you click 👍 or 👎 on a chat answer, we store four things in a Cloudflare D1 database:
- The rating (
1for up,-1for down). - The question you typed.
- The answer the AI gave.
- A timestamp.
We don’t store your name, email, IP address, or any identifier that could link the row back to you — because we never collect those in the first place. Feedback exists so we can find where the AI gives bad answers and improve the underlying writings or retrieval.
Aggregate web analytics. We run a self-hosted
Plausible instance at
plausible.ljs.app. Plausible is cookie-free and does not build
cross-site user profiles. We see page-view counts, referrers, and
scroll-depth percentages — aggregated, not per-person. We use this to
figure out which articles and journey paths people actually read, so
we know what to polish.
What we don’t collect
- No accounts. There is no login. There is nothing to log in to.
- No cookies for tracking. The site does not set any cookies at all for analytics or advertising.
- No chat transcripts beyond the feedback rows above. Questions you ask the chat flow through a Cloudflare Worker to Cloudflare AI Search and the answer streams back. Nothing is written to a database unless you explicitly rate the answer.
- No third-party trackers. No Google Analytics, Meta Pixel, TikTok, Segment, anything.
- No selling, no sharing. We don’t sell data. There is no data to sell.
Where things live
Everything — the static site, the Worker, the D1 feedback database, and the AI Search instance — runs on Cloudflare, which operates from the United States. Cloudflare’s own privacy terms apply at the infrastructure layer.
Who sees it
The maintainers of this site, for the purpose of improving answers and fixing gaps. That’s it.
Beta
This site is in beta. If how we handle data changes materially, this page changes with it. The date at the bottom is the last update.
Last updated: 2026-04-23.