Archetype and Battle Rhythm

The title of this article can sound a bit ominous, but it has real meaning in the world of Enterprise AI and agentic systems. It’s about a massive shift away from ad-hoc, bespoke AI experiments and into purpose-driven, deliberate execution with a pattern that establishes what the military calls a “Battle Rhythm.” I was introduced to this concept by leaders from the US Navy, and to be honest, it was shockingly enlightening.

I’ve had the pleasure of working closely with Mike Varney, a former US Nuclear Submarine Commander, and Cam Fulrath, a former US Navy Fighter Pilot for several years. Both have command officer experience that translates directly into the enterprise AI world we live in. Their stories are far more fascinating than my day-to-day dealing with code and hyperscalers, so much so that they quickly drew me in. Every time we talk, there’s a new story and a new learning.

It’s humbling, to say the least.

This article pulls on one particular thread from our conversations on how Command and Control translates into enterprise AI, and that’s Battle Rhythm. In order for a submarine to execute properly, everything is carefully trained and choreographed. Everyone has their position, job, and role to play, and they play it out in perfect sync. Even the specific “litany” used in communications is mission-critical to eliminate ambiguity and ensure flawless execution. Training, certification, credentials, and re-qualification are constant. That whole disciplined operational cadence is the Battle Rhythm.

This matters a lot more with agentic AI than it ever did with GenAI. A chatbot could give you a bad answer, but an agent can make a bad decision, trigger a command, access sensitive data, call another agent and actually do something. In cybersecurity terms, you now have to worry about hijacking, misuse, privilege abuse, cascading failures and agents going rogue.

So how does this play out in the white-collar world of the enterprise? Simply put, rhythm is critically important if you want things to actually hum along. The same discipline applied within the Navy is just as relevant to the processes, procedures, and rigor that every regulated enterprise has to live by. If you want trust in the business, you have to establish an A-level process. And that 100% applies to your enterprise AI initiatives.

To be fair, a lot of these terms like litany, certification, credentials, re-qualification and A-level process are coming right out of the conversation with Mike and Cam. They all make sense. The key is translating them into our everyday business lives.

We’ve all seen organizations experiment. Some do it to innovate, some to “test the waters,” and let’s be real, most are just trying to keep up with the Joneses. These experiments exploded over the past couple of years with GenAI, chatbots, and now agents. But they’re fraught with real dangers as they’re extremely brittle, lack the security and privacy enterprises actually need, require constant babysitting … and there’s a massive problem with scalable repeatability. In other words, there’s no Battle Rhythm.

So how do you actually get a Battle Rhythm in enterprise AI? Is that the million-dollar question or the billion-dollar question?

There are a few key pieces to making it happen, and one of the biggest is the Archetype. Now, most folks are going to say, WTF is an Archetype?

Simply put, it’s a model for how you do something. And I’m not talking about an AI machine learning model for anyone stuck way too deep in the AI weeds. I mean a pattern, a repeatable operational pattern. In the world of AI, the Archetype is the key to deterministic control and predictable execution across workflows that tap into fundamentally probabilistic models.

In a past article, I talked about Situational Decisioning in AI, where an intelligent system has to decide what to do next based on context and situational awareness. From an executive standpoint, that’s inherently nerve-wracking when you’re looking for deterministic, predictable outcomes in business. I also mentioned in the article that you need business rules on top of it, but what I didn’t state is that what you need is actually bigger than just rules. It’s Archetypes.

At CharliAI, we rely on Archetypes as the bounding boxes around workflows by defining how they should execute as well as what the workflows MUST execute regardless. The Situational Decisioning runs strictly within the bounds of the Archetype and cannot escape the pattern. Under this method, we’ve effectively created a fixed pattern of execution while leaving sufficient flexibility for situational decisioning, conditional processing, and variation in outcomes, as long as they meet the strict guidance established by the Archetype.

Under the covers, Archetypes are more than just rules, they are symbolic models, and they are critical to the neurosymbolic processing handled by our Forensic Control Plane. They are also mission-critical to the instrumentation, collection of evidence, and policy enforcement at runtime. Our customers can build massively flexible, adaptive and powerful agentic workflows on top of these Archetypes while sleeping comfortably knowing that the workflow execution can’t drift, can’t decay, and can’t go silently off into the wild.

The Archetype isn’t just defining the steps in the workflow. It defines the operating envelope: what has to happen, what can happen, under what conditions, and where the agent’s authority ends.

That is how the Archetype creates your Battle Rhythm.