Agentic AI:The Next Rung

Image by Midjourney

I’m sitting in the middle of an agentic AI course right now. And I’ll be honest. I came in expecting it to be challenging.

It’s been more than that. The process is complex, the technology is humbling, and “I’m sure it’s just me” has become my unofficial mantra. (Loudly. Who are we kidding.)

But I’ve felt this before.

There is a through line here that I keep coming back to, one I touched on in my last post about perspective and career pivots. The feeling of being in over your head is almost always temporary. The growth it produces is not.

Two years ago, I was terrified to write my first meta-prompt (a prompt designed to generate other prompts). It felt impossibly technical. Like everyone else in the room had a map I hadn’t been given. I wasn’t sure I’d ever move through this work with any real confidence.

And then, somewhere between the late nights and the small wins, I did. Not perfectly. But with grace. With efficiency. With a sense of what I was doing and why.

That’s what I keep coming back to as Sara Davison and Tyler Fisk walk us through How To Scale A Business With AI & Agentic Workflows. The overwhelm is real. And it is also not permanent.

The Prompt Era Was Just the Warm-Up

We spent 2023 and 2024 learning how to talk to AI. We got better at asking. We built Custom GPTs (personalized AI workflow tools), created prompt libraries, and trained our teams to stop using ChatGPT like a search engine with better manners. That was real progress.

It was also just the warm-up.

Agentic AI, systems that can plan, reason, and take sequential actions across multiple tools without constant human input, changes what a small, stretched team can actually accomplish. Think about what that means in practice. A lean marketing team today manually bridges every step: research, drafting, optimization, scheduling, reporting. Each handoff costs time a lot of teams simply do not have.

Agentic AI removes the manual bridge. The system plans the workflow, executes across platforms, and surfaces the outcome. The human sets the goal, reviews the result, and moves on to the work that actually requires their judgment.

For organizations that are strapped for headcount, that is not a threat. That is a lifeline.

The prompt era gave people something they don't always recognize. The ability to think in instructions and provide context. That's not a beginner skill. That's the foundational skill of agentic AI. Every agent in a system is executing on someone's clarity. The teams that built that muscle early are the ones who are moving the fastest now. Wade Murley, Resident, AI Build Lab LinkedIn

Image courtesy of Claude (Claude!)

Three Rungs. Enormous Opportunity at Every Level.

I’ve been working through a framework I like to call the Autonomy Ladder (yes, a bit buzz wordy, but you get the picture).

Rung 1: Prompt-Dependent AI. You ask, it answers. The AI is reactive. Still enormously useful. Most teams are here, and there is nothing wrong with that, as long as you know the next rung exists.

Rung 2: Workflow AI. AI executes multi-step sequences inside a defined process. Tools like Zapier connected to APIs, Custom GPTs chained together. You direct less; you configure more. One person starts doing the work of three.

Rung 3: Agentic AI. The system reasons about the goal, selects its own tools, adjusts mid-task, and acts with minimal human prompting (but a human-in-the-loop for safety and quality). This is where a team of five starts operating with the throughput of a team of fifteen.

The good news? The ladder is not about replacing people at any rung. It is about amplifying what each person can own.

This Is the Moment for Leaner Teams to Lead

Here is the reframe that I keep coming back to: the organizations that will move fastest into agentic AI are not the biggest ones. They are the most intentional ones.

Large enterprises move slowly. They have legacy systems, approval layers, and organizational inertia that makes transformation genuinely hard. Smaller, leaner teams, the ones who have been doing more with less for years, are actually positioned to move first.

If your team has ever wished you had a dedicated SEO specialist, a content coordinator, a campaign analyst, and a distribution manager, but the budget was never there? Agentic AI starts to fill that gap. Not by pretending to be a person, but by handling the sequenced, repeatable work so your actual people can focus on the strategy and relationships that move the business.

That is not a job going away. That is a team finally getting the support it always needed.

Agentic AI & What Your Team Should Do Now

You do not need to wait until you fully understand agentic AI to start benefiting from it. Three moves that matter right now:

Audit your workflows before you automate them. Map the manual steps from insight to published content. Not to hand them off immediately, but to spot where your team’s time is going that does not require human creativity or judgment. That is your first automation opportunity.

Build your documentation like it will brief an agent. Your brand voice guide, content standards, and approval criteria should be precise enough that an AI system could act on them consistently. The clarity you build now becomes your competitive advantage later.

Start with supervised autonomy. Tools like Cassidy and Claude’s tool-use capabilities allow human-in-the-loop configurations, meaning you stay in control while the system takes on more. Use that stage to learn. Adjust. Build trust in the process.

You have heard me say this before. Probably more than once. But it bears repeating here, because it lands differently for the teams who have been running on fumes: AI is not coming for your jobs. It is coming for the gap between what your team can do and what your team actually has the capacity to do.

Feeling overwhelmed by all this agentic talk? Remember, you don't have to master all of it.  You just have to learn enough of it to help your team take that next, first step.

And for those who have been running lean for years? This is your moment.

Ready to explore what agentic AI could unlock for your organization? Let’s talk.

Looking to learn? Check out classes and Lightning Lessons at AI Build Lab, taught by the amazing Tyler Fisk and Sara Davison.

Action over theory. Love it.

Finally, and I promise this is the last of the "announcements" (ha - I sound like a TV station), this article is a joint partnership between me and my AI collaborators, Claude and Midjourney.