Now

What I'm focused on right now. Last updated: August 2026.

Currently Exploring

  • FanoutFox patch. In progress. ChatGPT's search rebuild replaced the JSON that FanoutFox reads with a pipe language, so the fan-out card sits empty on fresh sessions until the extension speaks it. Building that patch now, as promised in the teardown.
  • Gemini network analysis. In progress. The last one in the set, and it needed a different method. Gemini answers from two layers, the app's own traffic and the grounding API, so the capture had to read both. The draft is written and I'm re-running probes before it goes out.
  • Image SEO follow-through. In progress. The GSC MCP image SEO release is out, 8 tools that read Google's image-search surface, with the launch post built on real client data. Now running the fix list on the six-figure page from that post and letting image_keyword_trends judge the work. The result publishes either way.
  • IndexNow experiment. In progress. Wired IndexNow into the deploy pipeline and seeded the full sitemap, 106 URLs. Now measuring whether the engines that listen actually crawl faster.
  • Charging AI crawlers with x402. Done. Built a proof of concept where this site charges AI agents a penny per page over HTTP 402, then watched Claude actually pay it. The settlements were real, not mocked. The writeup is drafted and needs screenshots and a recording before it ships.
  • Log file intelligence. Done. Built a real-time log analysis dashboard on a $5/mo Cloudflare Workers setup: verified bot detection by IP, AI-crawler tracking, crawl-budget views. Still need to write it up as a rebuild of my 2020 log analysis guide.
  • Internal linking. Done. Found an approach that works, embeddings plus a crawl, that beats what's out there. Will publish the writeup later.

Previously Explored

Tools and projects I've tested and formed a verdict on. Most recent month shown below. Older entries live in the archive.

August 2026

  • ChatGPT's new search language Kept web.run swapped its fan-out JSON for a pipe language in mid-August, and I read the new format line by line. Earlier in the month, the shortlist post showed ChatGPT often knows who's in the running before it searches. Both sets of facts live in the research tracker, re-verified as the engines change.
  • AI Conversation Query Classifier Kept Built and shipped alongside the AI Mode tracking guide. Paste your Search Console queries and a model separates ordinary keywords from the conversational queries leaking out of AI Mode.
  • Common Crawl Visibility Checker Kept Built and shipped. Reads 12 months of Common Crawl history for any domain or single page, dates the month a CCBot block appeared, and tells you whether a CDN template wrote it. The launch post covers why that archive decides whether you exist in AI training data.

View older months in the archive →

Living

  • Raising three kids, which remains the only job with no PTO and no performance reviews.
  • Trying to read more and spend less time on X. One of those is going well.