AI Search & Discoverability Engineering

Treating visibility as a data problem, not a guessing game

Google AI Overviews, ChatGPT, Perplexity, and Gemini answer questions directly now. Click-through is falling, citations are replacing rankings, and the gap that decides who gets cited is usually technical — malformed schema, an architecture crawlers can't traverse, entity signals that never resolve.

That gap is where I work. Most of what looks like a content problem turns out to be a data problem once you read the logs.

I take on this work through CitedBy, my consulting practice. If you're evaluating someone for a GEO, AEO, or technical SEO engagement, start there — services, case studies, and scoping all live on that site. This page is about the engineering behind it.

How I actually approach it

I write the code. I don't hand off a strategy deck and disappear.

  • Log file analysis — parsing raw server logs in Python and SQL to see what crawlers and AI agents actually fetch, how often, and what they abandon. Crawl budget is measurable; most audits never measure it.
  • Structured data as an API — Schema.org and JSON-LD treated as a machine-readable interface to your content, with entity relationships that resolve rather than a checklist of isolated blobs.
  • Indexation and architecture forensics — crawl diffs, canonical conflicts, orphaned depth, render-blocking paths that keep content out of the index in the first place.
  • Measurement that survives contact with a stakeholder — SQL-backed frameworks that track citation appearances and snippet coverage across discovery channels, wired to dashboards people will actually open.
  • Implementation in your codebase — Python, JavaScript, R, SQL. I work in the repo, alongside your engineers, with documentation so the capability stays in-house after I leave.

Twenty-five years of software engineering is the reason the recommendations are implementable. If you've worked with an SEO consultant who couldn't talk to your engineers, that's the difference.

Where the practice lives

GEO and AEO are young enough that the playbooks are still being written, which is most of why I find them interesting. I run that work commercially under CitedBy, which serves small and mid-sized businesses and agencies that need deep technical work white-labeled or named.

My own Insights here stay technical — implementation notes, CMS architecture, and the occasional post-mortem on something that broke.

Working with me

New client engagements go through CitedBy, including audits, implementation, and ongoing retainers. Request a free visibility audit →

For anything else — engineering roles, collaborations, speaking, or just comparing notes on how AI retrieval is behaving this month — get in touch directly.

Last Updated:
Contributors: Stephanie Fuda