GEO / AEO Specialist

Mike King Portfolio

Founder of iPullRank and author of the openly published "AI Search Manual," which frames GEO as "Relevance Engineering" — an engineering discipline combining embeddings, vector retrieval, and information-retrieval theory rather than a content checklist.

Relevance EngineeringVector RetrievalAI Search ManualEnterprise GEO

Location

Location: classified. Send a pigeon.

Skills

Relevance EngineeringVector RetrievalAI Search ManualEnterprise GEO

What makes it work

A breakdown of the choices that make this portfolio stand out.

1

A full open-access book instead of a lead-gen PDF

Publishing a 24-chapter manual in the open, rather than gating it behind an email form, positions the firm's thinking as genuinely worth reading rather than as a sales funnel.

2

A named reframe that argues against the field's own popular term

Explicitly arguing GEO should be understood as "Relevance Engineering," not a content trick, stakes out a distinct point of view in a crowded, buzzword-heavy new field.

3

Technical depth as the differentiator in a content-heavy field

Grounding the framework in retrieval mechanics (embeddings, vector search, passage-level scoring) signals a technical practitioner, not a content marketer relabeling old SEO advice.

What GEO / AEO Specialists can take from this

Specific, actionable tips to apply to your own portfolio — no generic advice.

  • If you have a substantial framework, consider publishing it fully in the open rather than gating it — it demonstrates confidence in the thinking.

  • In a new field full of shared buzzwords, stake out your own specific term or reframe and defend it.

  • Go a level deeper technically than most of the field is willing to — depth separates a real practitioner from a relabeled generalist.

Case studies

Mike King's GEO work reads like reverse engineering: a leaked Google API reference, a stack of patent applications, and a free tool that replays query fan-out with Gemini. The AI Search Manual turns all of it into a 24-chapter open playbook.

Case StudyadvancedOngoing; change log updated February 2026

The AI Search Manual

Relevance EngineeringQuery fan-outInformation retrievalGEO measurementPrompt recipes

An openly published manual iPullRank bills as "the official documentation for Relevance Engineering in AI search." It runs from how search shifted to conversations, through retrieval architecture and query fan-out, to measurement, team structure and ethics. Appendices add a glossary, an AI infrastructure tool index, retrieval-simulation prompt recipes and a citation tracker spreadsheet.

It shows he can turn a moving target into a structured, maintained body of work, and keep it current with a visible change log rather than letting it go stale.

  • 24 chapters, from lexical-to-neural retrieval to GEO analytics and entity attribution

  • Appendices ship working assets: prompt recipes for retrieval simulation, a citation tracker spreadsheet

  • Public change log: February 2026 update added chapters 21-24

Projects

TooladvancedIntroduced May 2025

Qforia

PythonStreamlitGemini APIPrompt engineering

A free query fan-out simulator. You enter a query and a paid Gemini API key, choose AI Overview or AI Mode, and it returns the synthetic sub-queries Google might run, each labeled with query type, user intent and the reasoning for its selection. King built the prompts from a 2023 Google Research paper on query expansion by prompting LLMs and a Google patent application on prompt-based query generation for diverse retrieval.

He took a behavior Google only described at a high level, grounded it in patents and published research, and shipped a tool others can run on their own queries.

  • Separate AI Overview and AI Mode fan-out modes; the model decides how many queries a prompt needs

  • Each sub-query tagged with type, user intent and selection reasoning

  • Method traced to a Google patent application and a May 2023 Google Research paper

ResearchadvancedPublished May 27, 2024

Secrets from the Algorithm: Google Search's Internal Engineering Documentation Has Leaked

API documentation analysisPatent researchDOJ trial testimonyInformation retrieval

A long-form analysis of accidentally published documentation for Google Search's internal Content Warehouse API. King cross-referenced the modules and attributes against earlier Google leaks, DOJ antitrust testimony and patent research for his book The Science of SEO, then walked through systems such as NavBoost and features such as siteAuthority.

It turned raw protobuf-style reference docs into claims a marketing team can act on, each tied to a named module or attribute, with the caveats stated up front.

  • 2,596 modules and 14,014 attributes reviewed from the leaked API reference

  • Click-signal module for NavBoost set against public testimony and patents

  • Explicit caveats: no scoring functions and no source code, only feature definitions

Ready to build your portfolio?

Follow the GEO / AEO Specialist roadmap — skills, projects, and timeline to get hired.

GEO / AEO Specialist Roadmap