PowerUp AI Mastermind — February 19, 2026
EigenThinking: the session that turned your cognitive fingerprint into a framework
“Your eigenthinking is the eigenvector of you — the direction that you naturally pull, the one that has the most variance across your thinking. That IS your intellectual property.” — Lou
This Week in 30 Seconds
- EigenThinking presentation — Lou introduced a framework for identifying cognitive eigenvectors: the recurring patterns in how you think that constitute your unique intellectual signature
- UP=IP — Unique Perspective equals Intellectual Property; the framework for turning observation patterns into branded methodologies
- Latent terrain cartography — navigating beyond the AI’s modal responses by deliberately exploring orthogonal perspectives in the latent space
- Reverse-engineering frameworks — asking AI to analyze your own conversations and extract the implicit methodology you’ve been using without knowing it
- ICH: Internalized Context Hub — grounding AI in your specific worldview and frameworks before asking it to build anything for you
- GEARS Alpha onboarding update — intake forms, ontology review process, GitHub walkthrough for skills distribution
- Elizabeth’s context-carryover protocol — a method for transferring full context between Claude threads without information loss
EigenThinking: Your Cognitive Fingerprint as IP
Lou opened with a presentation that reframed what intellectual property actually is for a knowledge entrepreneur. The starting point: stop trying to be novel in your content and start identifying what’s already distinctive in your thinking — the patterns you can’t help but fall into, the angles you always come back to, the questions you ask that others don’t.
He introduced the concept of cognitive eigenvectors — borrowed from linear algebra but applied to human cognition. In the mathematical sense, an eigenvector is the direction a transformation naturally preserves or amplifies. Applied to thinking: your eigenthinking is the direction your mind consistently pulls toward, across all domains, regardless of the topic at hand. It’s what makes your analysis recognizable even when you’re discussing an unfamiliar subject.
The claim that landed hardest: this isn’t something you develop — you already have it. The work is excavation, not construction. Your cognitive fingerprint is already present in everything you’ve written, every decision you’ve made, every conversation you’ve had. The task is to get AI to help you see it.
“If someone told you to remove your personality from your thinking, you’d be left with something generic that sounds like everyone else. What’s left when you don’t remove it — that’s your eigenthinking.” — Lou
Deep Dive: Insight - EigenThinking — Turn Your Cognitive Fingerprint Into Intellectual Property — The complete framework for identifying your cognitive eigenvectors and turning them into a named methodology.
💡 What This Means for You
The next time you catch yourself thinking “that’s interesting” or “that’s not quite right,” pause and ask: what lens am I using to make that judgment? If you can name the lens, you can brand it. Your distinctive way of seeing is worth more than any specific thing you’ve seen.
UP=IP: Unique Perspective as Intellectual Property
Lou formalized the implication of eigenthinking with a simple equation: Unique Perspective = Intellectual Property. The argument: in a world where AI can produce competent analysis on any topic, the only thing that remains scarce is the specific angle you bring — the questions you ask, the connections you make, the frameworks you apply.
This isn’t about originality in the sense of having never-before-seen ideas. It’s about the consistent, recognizable, hard-to-replicate way you approach problems. A coach who always asks about energy before strategy has a unique perspective. A consultant who always starts with the constraint before the solution has a unique perspective. The perspective is the IP.
The group discussed what this means practically: you need to name your perspective, document how it works, and train AI on it before asking it to build anything for you. An AI that doesn’t know your eigenthinking will produce generic output. An AI that does know it will produce work that sounds like you.
Dirk immediately connected this to the GEARS project: if you build a branded framework from your eigenthinking, that framework can become a GEARS category, which feeds into the content generation system, which makes the framework findable in AI search — a compounding loop from perspective to IP to GEO authority.
💡 What This Means for You
Think of three situations where your instinct led to a good outcome that someone else wouldn’t have seen. What’s the common thread? That thread is the beginning of a framework. Name it. Even a rough name gives you something to build around.
Latent Terrain Cartography
Lou introduced a more experimental concept: deliberately navigating beyond the AI’s most probable responses to find the useful territory in its latent space. The premise: AI models have a “modal” response to most queries — the most statistically likely answer given training data. Modal responses are competent but rarely distinctive.
Latent terrain cartography involves prompting the AI to explore orthogonal perspectives — deliberately off-modal directions that might surface non-obvious connections. The technique: after getting the AI’s standard answer, ask it to approach the question from a perspective that is the opposite of its initial framing, or from a framework the AI doesn’t typically associate with the domain.
Lou positioned this as an advanced technique for people who’ve already mastered standard prompting. The payoff isn’t reliability — it’s surprise. You’re using the AI to access parts of its training data that don’t usually surface in normal conversation.
The group was intrigued but noted the practical challenge: you don’t always know when you’ve found something genuinely novel versus something that merely sounds unusual. Lou acknowledged this and suggested treating latent terrain exploration as an input to your thinking, not an output to publish directly. The human judgment step — deciding whether the orthogonal perspective is actually interesting — can’t be skipped.
💡 What This Means for You
Pick a topic you know well. Get the AI’s standard answer. Then ask: “Now approach this from the opposite direction — what would someone who fundamentally disagreed with that framing say? What might they be right about?” See if anything in that response surprises you.
Reverse-Engineering Frameworks from Your Own Conversations
Lou demonstrated a technique that several members immediately recognized as overdue: asking AI to analyze your own conversations and extract the implicit methodology you’ve been using without realizing it. The setup is simple — paste a significant conversation or document into Claude, and ask it to identify the recurring frameworks, decision criteria, and analytical patterns present in your work.
What comes back is a mirror. The AI can see patterns you’re too close to notice, name frameworks you’ve been applying implicitly, and describe your reasoning architecture in terms you can then use to train other AI tools.
The group discussed where this is most useful: long coaching conversations, extensive email threads, recorded sales calls, anything where you’ve been doing high-quality thinking that hasn’t been explicitly codified. The AI doesn’t create the methodology — it makes visible what was already there.
Deep Dive: Insight - Ask AI to Reverse-Engineer Your Conversation to Recover Hidden Frameworks — How to use AI as a mirror to surface the implicit methodologies already present in your work.
💡 What This Means for You
Find your best work from the last six months — a client email you’re proud of, a coaching session that went exceptionally well, a presentation that landed. Paste it into Claude and ask: “What recurring frameworks and decision criteria are present in this work? What does it reveal about how I think about this domain?” Read what comes back with curiosity, not skepticism.
ICH: Grounding AI in Your Internalized Context
A brief but structurally important concept that came up in discussion: the ICH, or Internalized Context Hub. The idea is to give AI a comprehensive document that describes your worldview, your frameworks, your client archetypes, and your methodologies before asking it to do anything substantial. Without this, the AI is building on generic foundations. With it, it’s building on yours.
Lou referenced this in the context of eigenthinking: once you’ve identified your cognitive eigenvectors, you put them in your ICH. The ICH is the interface between your perspective and the AI’s capabilities.
Deep Dive: Insight - Ground AI in Your ICH Before Asking It to Build Anything (from chat) — Why the ICH is the infrastructure layer that makes all other AI work more coherent.
GEARS Alpha Onboarding Update
Lou shifted gears (literally) in the second half of the session to update members on the GEARS Alpha timeline and process. Key updates:
- Most NDAs are signed; intake forms are coming in from four or five participants so far
- The onboarding process: intake forms are read file-by-file, an ontology is generated, Lou sends a client-facing report for review, then the client confirms or corrects before schema is finalized
- Context memory issue: early versions ran out of context when intake forms got too large; Lou re-architected to process file-by-file, resolving the problem
- For members without websites: use Claude (Opus is recommended, or the newly-available Sonnet 4.6 which has caught up on web design tasks) to build a simple placeholder site in a day — Donald built his in one pass in about an hour
- API cost transparency: Lou will absorb costs up to 50-$100 may require shared cost
- Licensing discussion: Lou raised the possibility of offering non-commercial source code licenses to mastermind members, with pricing likely in the 10,000 range for active use cases; a $500 training-day option was also floated for those who want education without the code
Donald’s Opus-generated website moment deserved its own spotlight. Having tried to build a comparable site the “old way” over a week of iterations, he completed the entire process in roughly an hour with Opus — including troubleshooting a mobile layout issue in two or three prompts. Lou immediately offered Donald’s site as a template: “Get the link to this site, give it to your Opus, and say, build a site like this for me in blue and silver, and have it interview you for the bits of information.”
💡 What This Means for You
If you’ve been putting off updating or building your website because it feels overwhelming, this is the signal you needed. The barrier is now “one hour with Opus.” Start with a placeholder: three pages, your brand, your audience, your transformation. That’s enough to begin building GEO authority.
GitHub Skills Repository: Getting Started
Lou walked the group through the AIMM GitHub repository — how to clone it, how to add skills, and how to keep it synchronized. For members new to Git, he demonstrated the VS Code approach: install VS Code, clone the repository via the interface (no command line required), add files using the Explorer panel, commit and push when ready.
The critical workflow reminder: commit and push makes your changes available to everyone. Fetch or pull syncs others’ changes to your local copy. If you accidentally delete something, “revert all changes” in Claude or VS Code restores to the last commit.
Two extensions Lou has installed in VS Code that make this seamless: Claude Code for VS Code, and Codex. Both give you an AI assistant that’s context-aware of which repository and file you’re working on.
Elizabeth shared a contribution: a context-carryover protocol she’s been refining that transfers full context from one Claude thread to a new thread without information loss. Lou’s reaction: “I need that yesterday.” She offered to share it via the GitHub repository.
💡 What This Means for You
If you haven’t accepted your AIMM GitHub invitation yet, do it this week. Clone the repository. Even just downloading the existing skills and reading them is worth an hour of your time.
Community Corner
Donald Kihenja’s Opus-generated website story is the kind of concrete “this is what’s now possible” data point that resets everyone’s calibration. His transition from “this took me a week last time” to “this took an hour with Opus” is something every member should hold onto when they’re procrastinating technical tasks.
Elizabeth Stief offered to share her context-carryover protocol with the group — a contribution that prompted genuine excitement. Lou’s “I need that yesterday” reaction said everything about how useful a reliable thread-to-thread context transfer would be for daily AI work.
Dirk Ohlmeier is now using skills regularly and has hired a Python developer — Claude is being used as the interface and communication layer between Dirk and the developer, which Lou flagged as a genuinely smart architecture for working with technical contractors.
Links Shared in Chat
- AIMM GitHub Skills Repository — private repository; access via GitHub invitation from Lou
Try This Before Next Session
Start your eigenthinking excavation.
- Find three pieces of your best thinking from the past year — written work, recordings, significant decisions.
- Paste each one into Claude separately. Ask: “What analytical frameworks and decision criteria are present in this work? What does this reveal about how the author thinks about this domain?”
- Read all three responses and look for what’s common across them. What patterns appear in all three? That’s the beginning of your eigenthinking map.
- Draft a one-paragraph description of your cognitive fingerprint — not your area of expertise, but your way of thinking about problems in that area.
Bring it to next session. We’ll compare notes.
Open Threads
- What’s the practical workflow for turning a documented eigenthinking pattern into a named, trainable framework — something you can put in your ICH and actually use?
- At what point does GEARS content generation replace manual article writing, and what does that transition look like in practice for someone just starting?
- Elizabeth’s context-carryover protocol — what exactly does it transfer, and are there limits to what can survive a thread restart?
- Is latent terrain exploration a reliable technique, or is it more dependent on the specific model and prompt formulation than it appears?
Next session: February 26, 2026
Derived Artifacts
- context-carryover (Context-Carryover Protocol — Elizabeth Stief)
- cognitive-fingerprint (Cognitive Fingerprint — Lou’s Eigenthinking UP→IP framework)
- SKILL (Eigenthinking Skill — Lou’s UP→IP framework)