You See It. You Build It.

During a stretch of contract limbo — that strange, fertile in-between space that most leaders either waste or weaponize — I came across a LinkedIn post that caught my full attention. A contact of a contact had built a fully interactive AI version of himself: capable of representing his expertise, fielding client questions, and projecting professional presence without him physically being available. Powered by Claude. Polished. Purposeful. The kind of thing that makes you sit up straighter. I looked at it for a long moment, then thought the thing I always think when I see something that works: I want to build that.

"That instinct — I see it, I want to understand it, I'm going to build it — is what digital transformation actually looks like from the inside."

I want to be honest about what I brought to this project. My background is in creating, leading, and delivering transformation — building the vision, assembling the capability, driving the outcome. I am not someone who has spent the last decade deep in code. What I have spent that time doing is understanding how technology creates value, how to frame a problem so a solution becomes obvious, and how to move fast when the window is open. That combination, it turns out, is exactly what AI-assisted building rewards.

I set myself a deliberate brief. I wanted an AI-powered executive personal assistant — something that could manage my calendar, represent me professionally to inbound client interest, and handle the ambient communication overhead that quietly drains strategic thinking time. My constraints were equally deliberate: no voice cloning, no AI-generated likeness. My professional presence, without my face or voice in the room. The outcome defined. The method open.

The process was a masterclass in directed iteration. I took a screenshot of the app that had inspired me, fed it to my LLM of choice, and used it as the brief. From there, the work was not writing code — it was thinking clearly. Articulating the outcome. Describing the logic. Catching where the model misread the intent and redirecting it firmly. The role I played was the same role I play on any transformation initiative: clear vision, sharp feedback, relentless forward motion.

One thing I'll say directly for any leader considering this kind of experiment: the models are not equal, and they are not infallible. I worked across both Claude and ChatGPT throughout the build, and found real differences in how each handles nuanced instruction, architectural reasoning, and the kind of iterative refinement that separates a working product from a demo. Getting to the standard I needed — something genuinely fit to represent me professionally — required persistence, precise correction, and the willingness to reframe the brief entirely when the output wasn't landing. The vision has to stay sharp. The models will follow a clear leader.

"The constraint was never capability. It was always clarity — of outcome, of intent, of what 'good' actually looks like."

What I ended up with was a working, integrated system built in a fraction of the time a conventional development cycle would have required. Not because the problems were trivial — several weren't — but because the gap between having a clear vision and having a working product has genuinely, materially closed. The person who inspired this brought deep academic and domain expertise. I brought a different kind of intelligence: the ability to see an outcome, frame it precisely, and drive toward it without getting lost in the implementation weeds. Both paths produced working software.

This is what I think leaders in our space need to sit with. The conversation about AI in the enterprise has spent too long framed around enablement — what AI can do for technical teams, for developers, for data scientists. The more interesting conversation is about what it does for people who have always known exactly what they wanted to build, but depended on others to build it. That dependency is dissolving. The gap between vision and execution is closing faster than most organisations have planned for.

If you're in a transition right now, use it. If you've had an idea sitting in a slide deck for two years because you never had the right technical resource available, pull it out. Take a screenshot of something that impressed you. Start the conversation. The model will meet you where your clarity is — and if your clarity is strong, you'll be surprised how far you get.