03 · writing

Engineering notes.

Practical writing on product systems, structured data, AI retrieval, and the controls that keep software reliable in production.

featured argument01
2026-08-0411 min read

After 500+ A/B tests, I started testing the interface no human sees

AI visibility depends on a second interface assembled by crawlers, answer engines, and agents. I call it the shadow interface, and propose a testable funnel for evaluating it.

human / machinetwo interfaces · one product
human
01 see
02 understand
03 trust
04 act
machine
01 fetch
02 render
03 resolve
04 retrieve
05 verify
06 refer
Read the field note
archive

The notebook.

Implementation notes, field observations, and systems-level arguments.

  1. 02How AI answer engines actually resolve your @graphChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews each parse JSON-LD differently. The deltas matter. Here's what I've measured running the same @graph through all five.
  2. 0310 JSON-LD bugs I find in every commerce audit (and what they cost)Field-level JSON-LD bugs that fail silently across 45K+ PDP catalogs. Each one mapped to the rich result it kills, the AI engine that drops you, and the dollar cost I model for clients.
  3. 04Schema decays. Here's the half-life I've measured.Production schema goes invisible to AI engines on its own, with no obvious failure. Here's the regression timeline I've measured across client engagements, and why CI gates aren't optional.
04contact

Looking for an engineer who can own the product and strengthen the platform?

I am considering Senior, Staff, and Principal roles where I can ship across the stack, improve the frontend platform, and build reliable AI-assisted engineering systems.

direct email · no scheduling gate
start a direct conversation
your message goes directly to me · no scheduling gate// legitimate interest (GDPR Art. 6(1)(f)) — you requested the contact. privacy policy