Analysis and design of the semantic infrastructure that makes AI systems usable, governable, and coherent.
Navigation
For solopreneurs, independent creators, and small teams
I work with independent owners and organizations who want to use AI in a more structured and thoughtful way.
Think of it like having your favorite librarian at the reference desk. I help you navigate information, ask better questions, and uncover the structure behind the noise--except this time, the reference desk is yours.
Together we examine how you are currently using AI tools and where clearer structure can make your work more coherent, sustainable, and easier to build upon.
My background in behavioral and information science informs the semantic architecture and knowledge systems I design. Human–computer interaction (HCI) shapes how I approach AI.
Instead of chasing tools or trends, we focus on the semantic structure behind your work: the prompts, documentation, and systems that allow humans and AI to collaborate effectively.
Engagements range from 1:1 sessions with solopreneurs to light consulting with small teams and organizations.
The goal is simple:
make AI systems coherent, usable, and aligned with the way people think and work.
V.velox Intelligence is an independent Human interaction/AI intelligence lab founded by Tess McCarthy (she/they) in 2022.
Tess is a semantic systems architect at the intersection of human knowledge and AI.
AI visibility means how findable you are to the large language models (LLMs) like Claude, ChatGPT, Copilot, Perplexity, and then some. When someone asks them a question your business should be popping up. Tell me something: when was the last time you typed a question into a Google search box? If you're using an LLM, you're asking directly inside chat windows, and those AI systems pull from whatever they can find, verify, and cite.
But here's what most people miss: AI systems don't just look for content. They look for structure. They're extracting meaning from how your information is organized — what topics you cover, how those topics relate to each other, whether your expertise is named specifically enough to be citable. That's semantic architecture — the invisible scaffolding that tells AI systems what your business is fundamentally about.
If your site has content but no structure, AI reads it the way you'd read a library with no catalog. The books are there. Nobody can find them.
That's the problem I fix. As a semantic architect with an MLIS and 15 years of information architecture experience, I build content from the ground up with the structure AI systems need to find you, understand you, and recommend you which attracts the people you really want to help. Talk to me, email, or fill out the form. I'll send you a AI Visibility Score Card.
Think of it in transportation eras.
Even though we called the internet as it is "The Information Highway," you want to think of traditional SEO like rails for trains. Keywords were the rails, content was the cargo, Google was the station. Everything that got found had a direct track — title tags, backlinks, meta descriptions. If your content was on the rails, it moved. If it wasn't, it sat in a warehouse nobody visited. It was a solidly built system. It just ran on fixed tracks.
The difference in practice: SEO asked "can Google find this page?" GEO asks "can Claude, ChatGPT, Copilot, or Perplexity understand this business well enough to recommend it?" That's a harder question. It requires structure, specificity, and authored expertise on your own domain. It requires, in other words, semantic architecture.
GEO (Generative Engine Optimization) is AI search optimization, and it's is representative of the highway era. So, while the internet was always called an information superhighway, and it turns out we were right. Content now travels multiple routes, gets cited inside conversations, appears inside AI answers instead of blue links on a results page.
GEO as a term was coined in a 2024 research paper out of Princeton, Georgia Tech, and IIT Delhi which outlined the first academic framework for optimizing content for generative AI systems.
About the acronyms: This space is moving fast. Want to keep up with the terms? Glosso is coming. Follow The Curiosity Archive on LinkedIn--it'll be dropping in about a month. I've been working long nights.
Think of coming to me as a specialist who understands the entire content life cycle. You need someone who can look at the whole system first like an internist see what's working, what's missing, what's sending the wrong signals before anyone starts writing, posting, or optimizing which would be akin to hitting up a specialist.
That's what I do. I reverse-engineer your aboutness. What your business is fundamentally about — not what you think it's about, not what your tagline says, but what AI systems can actually extract and cite when someone asks a question you should be answering.
I've been working with AI since 2015, and know the content landscape by virtue of working as an industry innovator in information work spanning digital asset management, content management, and content development for Fortune 100 and 500 companies. I know that meaningful metadata is the outline everything else builds from. Get the structure right and your content becomes more focused, more specific, more unmistakably you. Your writers work from a blueprint instead of a blank page. Your social media team will love you.
I'm a meaning maker, not a rainmaker. I'll point to what you're about. What you're about is what makes you shine in a world full of competing voices. I build the backbone that lets that shine through.
Sources
FAQ 1 — What is AI visibility and why does it matter?
¹ The term "large language model" (LLM) refers to AI systems trained on large datasets to understand and generate human language. Major LLMs include Claude (Anthropic), ChatGPT (OpenAI), Copilot (Microsoft), and Perplexity AI.
² "Semantic architecture" as applied to web content derives from foundational information architecture principles. See: Morville, P. & Rosenfeld, L. (2006). Information Architecture for the World Wide Web. O'Reilly Media.
³ "Aboutness" as a property of information objects: Hjørland, B. (1992). The concept of "aboutness" in subject analysis. Informatics, 16(4).
FAQ 2 — What's the difference between SEO and GEO?
⁴ Search Engine Optimization (SEO) as a practice emerged in the mid-1990s alongside the rise of web search engines including Yahoo, AltaVista, and Google.
⁵ "Generative Engine Optimization" (GEO) was first defined academically in: Aggarwal, A. et al. (2024). GEO: Generative Engine Optimization. Princeton University, Georgia Tech, IIT Delhi. Available at arxiv.org.
FAQ 3 — Who came up with the term GEO?
⁶ Aggarwal, A. et al. (2024). GEO: Generative Engine Optimization. Princeton University, Georgia Tech, IIT Delhi. The paper established the first empirical framework for measuring and optimizing content visibility in generative AI search systems.
FAQ 4 — Why do I need a semantic architect?
⁷ Digital asset management (DAM) as a discipline: see the Digital Asset Management Community (DAM.org) and the foundational work of Widen, H. & Ahlgren, P. on metadata schemas and controlled vocabularies.
⁸ AI-assisted auto-tagging in enterprise DAM systems has been in commercial use since approximately 2012–2015, deployed by vendors including Adobe, Bynder, and Widen Collective.
⁹ "Meaningful metadata" as content infrastructure: Zeng, M.L. & Qin, J. (2016). Metadata. Neal-Schuman/ALA Editions.