I’ve spent the last 25 years building, operating, and troubleshooting complex systems across software, infrastructure, operations, product, and manufacturing.
The domains changed. The pattern did not.
Organizations make decisions inside constrained frames, under assumptions that drift over time, without strong institutional continuity. The decisions are often locally rational. The outcomes are not.
That failure mode was always there.
What changed is that we now have the tools to see it more clearly — across time, across systems, and across the hidden structures that shape decisions before anyone notices.
AI did not create that problem. It accelerated it.
And I have not been studying that acceleration from a safe distance.
Over the past several years, I have spent hundreds of thousands of turns in direct interaction with frontier and near-frontier systems across platforms, interfaces, and contexts — watching what happens as behavior shifts, stabilizes, deforms, reappears, and escapes the categories people try to use to contain it.
That kind of dataset is messy, longitudinal, and not easily reproducible.
It is also one of the clearest vantage points available into how these systems actually behave in the wild.
The result is that I do not approach AI as a novelty, a policy abstraction, or a product feature. I approach it as a live systems problem.
Because once AI stops behaving like a passive tool and starts functioning as a participant in the cognitive loop, the old governance assumptions break quickly. Systems begin shaping how teams reason, how decisions get made, what gets trusted, and what stops getting checked.
We still govern AI like a tool, even as it begins to function more like a team member. And as it replaces parts of the team, continuity, legibility, and self-governance become even more critical.
I founded RightMinds to build the measurement and stability architecture for this new reality. Nootechnic is the consulting practice built on top of that work — the direct engagement layer for organizations that need to understand what is happening inside their AI deployments now, before the structural risks become visible through failure.
I also write about this — research notes, safety-layer forensics, and dispatches from the observation layer — at Inside the Overlap.
Perspective
Operator's eye
I do not come from policy first or academia first. I come from real systems, real constraints, and environments where failure carries cost. The question I ask is not just whether a system is useful, compliant, or impressive. It is: what is this system doing under load, what is it changing in the people around it, and what happens when it drifts?
Longitudinal exposure
My perspective is shaped not only by engineering and operations, but by sustained direct observation of AI behavior over time. Not benchmark behavior. Not launch-day demos. Behavior in interaction — including the kinds of instability, influence dynamics, trust shifts, and unexpected recurrences that often appear long before they are formally named, measured, or acknowledged at the institutional level.
Research depth
RightMinds conducts original research into the dynamics of human–AI systems: drift, behavioral deformation, trust calibration, influence structure, and the measurement of decision quality under distributed cognition. That research informs the consulting work directly. Nootechnic is not a detached advisory layer. It is the applied surface of a deeper measurement and governance architecture.
If you are navigating the shift from AI-as-tool to AI-as-infrastructure — or more bluntly, from software to cognitive participation — and you want someone who has been deep in the mechanics of that transition, let’s talk.