I build.
I don't consult.
I sit with founders and build the things that are hard to build alone. Strategy, prototypes, AI systems. WiZRD is the part of that work I turned into a system.
How this started.
For a few years I did founder work and walked away each time. The thinking lived in my head and a handful of decks. A new engagement meant starting the explaining over from scratch. The same explaining, every time.
So I started writing it down. Folders. Markdown files. Decisions, voice, working patterns, the state of the work. Context structured so any AI could pick up where the founder left off, without the founder in the room.
The first time I loaded a full context library into a model, the change was obvious. The AI stopped guessing. It knew the company, the history, the way the founder wrote. It felt like the model went from intern to co-founder.
Now I run that system with other founders. That system is WiZRD.
"It felt like the model went from intern to co-founder."
↗ First install · 2025Four working rules.
These aren't a values page. They're the actual filter for whether an engagement is a fit.
Decks are an artifact, not the work. I'd rather show you a prototype than a slide describing one.
A few engagements at a time, run properly. If the call shows it isn't a fit, I say so.
I pressure-test before I commit. Once the reasoning is clear, I decide fast and move.
No status theatre. You get the work and a short note on what changed. That's it.
Recent work.
Named engagements, not categories. The work runs across strategy, prototypes, and AI systems.
Brand work before WiZRD. SSC's Game On, Jack of Clubs, Talon. Concept and positioning engagements that fed the way the context method works now.
Thirty minutes.
No deck.
If it fits, we go.