Training crawlers
—
Models learning from this research.
Bots scraping pages to train large language models. The lab’s findings become tomorrow’s answers.
Case Study — Applied LLM Systems
Applied LLM systems, for operators.
minnesota.ai is the home for Applied LLM Systems — the systems-level work of making AI discoverable, parseable, and citable for operators. Where daveedvalencia.com focuses on Chat Visibility specifically, minnesota.ai covers the full systems layer through controlled experiments, frameworks, and live data — all published in the open.
ChatGPT conversion
24.9%
documented in a controlled experiment
Foundations
3
structure / visibility / discoverability
Methodology
Open
all findings published
Crawler data
Live
updated every 24h
01 — The setup
Systems-level work, applied to live sites.
minnesota.ai runs Applied LLM Systems work on real domains for real operators — the systems-level engineering of making AI discoverable, parseable, and citable. Each experiment runs on a live site with real traffic, not a sandbox. Findings get published as research papers with live data, methodology, and conclusions in the open.
The three foundations — Structure, Visibility, Discoverability — emerged from this work. So did the 24.9% ChatGPT conversion finding. The same methodology that lives here powers the client engagements I run.
02 — The proof
The lab measures its own work.
AI returns to sites it has learned to trust. minnesota.ai is one of those sites — even between content drops, the crawler activity continues. Below is the live data from minnesota.ai in the last 24 hours.
AI search bots
—
Models indexing for retrieval.
Bots powering AI-driven search experiences. Pages get surfaced when someone asks an AI about AI visibility.
Live assistants
—
Models citing the research right now.
Bots triggered live by user prompts. The lab’s pages get pulled directly into AI conversations as they happen.
03 — The thesis
The next economy runs on agents.
AI systems are the new search layer. They route users to the businesses that can serve them. Your site is the infrastructure that decides whether AI sends users to you, or somewhere else.
minnesota.ai exists to figure out exactly how that routing works — what AI parses, what it trusts, what it cites. Every finding goes back into the methodology applied to client work, and back out onto the site as published research.
AI is the new search layer. Optimize for it.
Work with me →