For the past two years, I’ve been an enthusiastic, somewhat scattered AI experimenter. I have favorite tools. I’ve built clever little workflows. I’ve been invited to explain the difference between augmentation and automation in meetings and classes. And yet, until recently, I couldn’t see a clear path from me using AI well to my team using AI well at scale.
That gap is why I enrolled in the AMP/Momentum AI Operators Bootcamp.
The AMP course is led by Rachel Woods, founder of The AI Exchange. She was working in AI before ChatGPT — long enough to recognize the patterns of how AI scales (or doesn’t) inside organizations. The program runs six weeks, with a longer certification track on the back end, and distills those patterns into a framework, a master prompt structure, and a tool-agnostic playbook format. It isn’t a tool tour. It’s a way of thinking about AI as a team sport.
The piece I’d been missing
In the first week, we learned about the five roles needed to make AI work inside an organization:
- The Visionary — sets the goal and frames what success looks like.
- The Operator + Subject Matter Expert (SME) — maps the plan and defines what “good” looks like.
- The AI Implementor — builds it. Configures the tools, the data connections, the automations.
- The SME (review) — checks the output for accuracy and gives clear, structured feedback so the Implementor can iterate.
- Everyone else — actually uses what’s been built. As Rachel puts it, “adoption is the measure of success.”
Reading that list, I finally understood where I’d been getting stuck.
I had the Visionary part, mostly. I could be the SME on plenty of processes I wanted to improve. But I’m not a coder, and I have very little Implementor experience. I had been trying to do a five-role job alone — which, predictably, led to half-built GPTs and a vague sense that “AI in our workflow” was something I’d get to one day.
Naming the roles and responsibilities changed the conversation. It clarifies my job as the leader — to articulate the why and the what clearly — and opens a door for colleagues with deeper technical skill to bring their how and tell me when something won’t work. It also makes building AI capacity shared and teachable, which is what higher ed and other mission-driven organizations need right now.
Why the framework matters more than the prompts
Free prompt libraries are everywhere. Most of them are fragmented and don’t address the holistic, ongoing needs of a team trying to build something durable.
What AMP offers instead:
- A master prompt structure — markdown headings, ordered context, clear ask — that AI models actually use well. (I used it to draft the prompt for this very post.)
- A playbook format — tool- and subject-agnostic — for capturing and improving how repeatable work gets done.
- Practice. A lot of it. Last week, between videos, the live session, and homework, I spent about ten hours on playbooks, with feedback from a custom GPT the AMP team built. It was hard. It was worth it.
A framework, in my experience, is priceless. It tells you what’s important, where you’re headed, and the steps to get there—grounding AI work in practical strategic thinking techniques instead of scattered trials.
A short watch
If you want a feel for how Rachel thinks, Tiago Forte’s interview with her — The High-Paying AI Job Nobody Knows About (Yet) — is a good 45 minutes. AMP also offers a free intro sessions (follow on LinkedIn) and shorter four-week bootcamps if six weeks feels like too much right now.
An invitation
If any of this resonates—somewhere between solo AI experiments and figuring out how to bring a team along—you’re probably already searching for more effective leadership strategies for AI. I’ll be writing more as I move through the program.