From Additional Duty To AI Practitioner

AI integrator is not my official title. It started as an additional duty. Now it is what I do full time.

I figure out where AI can improve our products, save time, and make the planning process more effective.

That sounds straightforward.

It usually is not.

I plan military command post exercises for brigades and below.

If you have worked in that environment, you know how many moving pieces there are: orders, scenarios, injects, briefings, products, coordination, revisions, and deadlines.

There are a lot of places where AI can help, but only if it is applied carefully and in the right part of the workflow.

Figuring out where it fits is the job.

So how did I learn AI?

Mostly by doing.

My boss and I have been interested in AI since ChatGPT first became widely available. But I did not get serious about applying it to our work until 2024.

At first, I took short courses on prompting, AI basics, and how these tools work. My company has a Udemy account, so I started down the certification path too.

Then my boss came to me with requirements.

The need was immediate. We did not have months to wait for me to finish a course sequence and come back with a certificate. That does not mean certifications are not valuable. They are. It just means my situation required me to become useful quickly.

So I put the certification journey on hold and started learning how to build.

I used YouTube, short courses, documentation, trial and error, and a lot of hands-on work. There were plenty of mistakes, false starts, and do-overs. Most of the learning happened while trying to solve real planning problems, with real products, for real users.

Over time, my boss and I stopped being AI enthusiasts and started becoming AI practitioners. We have learned a lot about what AI can do, what it cannot do, and where human judgment still has to lead.

The best example so far is CADE: the Combined Arms Decision Exercise.

CADE is a tabletop staff exercise built to help partner-nation staffs develop decision-making, planning, and synchronization skills in a NATO-aligned context. It has now been through three iterations, and it has taught me more than any course could have about building practical AI-assisted training tools.

I plan to write a full series on how CADE was developed: the design process, the mistakes, the improvements, and the lessons learned along the way.

This is the starting point for that series.