Field Notes

Applied research on AI search optimization and AI agent development

Content Roadmaps in the Age of AI Search

Content Roadmaps in the Age of AI Search

Content roadmaps built for the AI search era start with retrieval fitness, not keyword volume.
Competitive Intelligence in AI Search

Competitive Intelligence in AI Search

Competitive intelligence in AI search is the practice of analyzing which brands get cited by
Wikipedia or Die: How to Claim Your Q-Node and Own LLM Entity Disambiguation

Wikipedia or Die: How to Claim Your Q-Node and Own LLM Entity Disambiguation

Wikipedia and Wikidata Q-nodes anchor entity disambiguation in large language models. When an LLM encounters
Embedding Visual and Audio Assets for LLM Retrieval: A Field Manual to Multimodal Immortality

Embedding Visual and Audio Assets for LLM Retrieval: A Field Manual to Multimodal Immortality

Multimodal LLM retrieval processes images, audio, and video alongside text. Embeddings convert visual and audio
Mentions > Links: A New "Ranking" System for AI Search

Mentions > Links: A New "Ranking" System for AI Search

In AI search, brand mentions across authoritative sources matter more than backlinks. LLMs do not
How to Use Endpoints to Drive LLM Citations

How to Use Endpoints to Drive LLM Citations

Endpoints give LLM retrieval systems structured access to your brand data. A dedicated JSON-LD fact
Use Public Repos to Pull Your Company into ChatGPT Answers

Use Public Repos to Pull Your Company into ChatGPT Answers

Public repositories on GitHub, Hugging Face, and Zenodo function as training-data pipelines that large language
The Top 10 AI SEO Agencies in 2025

The Top 10 AI SEO Agencies in 2025

AI SEO (also known as Answer Engine Optimization or Generative Engine Optimization): the art and science of engineering content that large language models (LLMs)—like ChatGPT, Google's SGE, and Bing Chat—can't resist citing.
The 2025 Perplexity Playbook: Sonar Ranking Factors

The 2025 Perplexity Playbook: Sonar Ranking Factors

Perplexity's Sonar model uses a retrieval-augmented generation pipeline that selects, ranks, and cites
Answer Shapes 101

Answer Shapes 101

Answer shapes are the discrete structural patterns that large language models prefer to ingest, extract,
Why Brand, Distribution, and Trust Are the Last Moats

Why Brand, Distribution, and Trust Are the Last Moats

Brand authority, distribution power, and trust capital are the only durable competitive advantages remaining in