AI Search Engineering
A Technical Guide to Generative Engine Optimization (GEO), AI Citations, and Agent-Ready Websites
A technical, evidence-first guide to Generative Engine Optimization. Instead of treating GEO as a collection of tricks, the book presents it as an engineering discipline built around a measurable pipeline — the same methodology the open-source GEO Optimizer audit engine implements.
- Author
- Juan Camilo Auriti
- Pages
- 992
- Language
- English
- Published
- 10 September 2026
Not ready to buy? Start free with the 160-page GEO Readiness Manual.
The pipeline the book is built around
Every chapter attaches to one stage of the path a page travels before it can be cited. Naming the stages is what makes GEO measurable instead of anecdotal.
-
01
Discovery
How answer engines find a URL exists at all.
-
02
Crawling
Crawler access, robots controls, and what actually gets fetched.
-
03
Machine understanding
Rendering, structured data, semantic HTML, entity signals.
-
04
Retrieval
Which passages get selected as candidate sources.
-
05
Answer generation
How content is used once it reaches the model.
-
06
Citation
Whether the source is named, and how to measure it over time.
-
07
Interaction
Machine-readable interfaces and agent-driven navigation.
-
08
Action
Preparing sites for agents that do things, not just read.
What it covers
- How modern AI answer engines discover and use web content
- Telling documented behaviour apart from research findings, observation, inference, and product-specific scoring
- Auditing crawler access, robots controls, rendering, structured data, semantic HTML, and entity signals
- Engineering content for extraction and citation without relying on arbitrary "AI-friendly" formulas
- Designing prompt research frameworks for measuring real AI visibility
- Tracking citations, source usage, answer presence, and visibility over time
- Building operational GEO workflows for teams, agencies, and technical SEO programmes
- Evaluating llms.txt, schema markup, and crawler directives without overstating what they can do
- Preparing websites for AI agents, machine interaction, and action-oriented interfaces
- Managing prompt injection, content integrity, and agent-related security risks
The appendices cover GEO audits, AI crawlers, JSON-LD templates, prompt research, citation tracking, evidence classification, CI/CD checks, client reporting, terminology, and technical references.
Who it is for
- Developers
- Technical SEOs
- Consultants
- Agencies
- Product teams
- Site owners
Written for people who want to understand what can actually be measured, implemented, and defended in AI search. No guarantees, no invented ranking factors, no hype presented as fact — the same standard the audit engine holds itself to.
Book and tool, same methodology
The book explains the reasoning; GEO Optimizer runs it. The eight-category rubric
described in print is still in the MIT-licensed engine (--score-version 1),
next to the nine-category v2 that is now the default — you can read the rules and disagree with them.
Questions about the book
How long is AI Search Engineering, and in which formats?
992 pages, in English, published on 10 September 2026 as a paperback (ISBN 979-8172934568) and a Kindle edition, both sold on Amazon.
Who wrote the book?
Juan Camilo Auriti, who also maintains GEO Optimizer, the MIT-licensed audit engine that GeoReady runs.
Does the book describe the same scoring as the GeoReady audit?
The book describes the eight-category rubric. The engine has since moved to a nine-category rubric (v2) as the default; the rubric in print is still available with geo audit --score-version 1. Both are open source under the MIT licence, so you can read the rules and disagree with them.
Is there a free way to start?
Yes. The GEO Readiness Manual is free (160 pages), and the GeoReady audit runs without an account. The book is the long-form version of the same method.