There's a gap forming inside agencies right now that nobody's posted a job listing for. It sits between the executives who know things need to change and the teams who are either resisting or just don't know where to start. And it's not a "Head of AI" role, because that sounds like someone who writes internal memos about prompt engineering and runs lunch-and-learns nobody wants to attend.
What I keep seeing, and what I think is going to become one of the most important positions in any creative shop over the next couple of years, is something closer to an internal operator who can move fluidly between strategy and execution. Someone who thinks in concepts but can also build the thing. The kind of person who understands the business problem, sees how the new tools can solve it in ways that weren't possible six months ago, and then actually makes it happen instead of writing a deck about it.
The gap is real and it's widening
Leadership sees the wave coming and knows they need to move, but their teams are either dragging their feet or actively pushing back because they think automation means their job disappears.
And I get it. If you're a mid-level designer or developer at an agency and your boss keeps talking about AI efficiency, you're hearing "we want to do what you do but cheaper." That framing is the problem. The agencies that are going to win figured out how to do work that wasn't even possible before, not how to do the same old work faster. Doing the same work faster doesn't get you anywhere different than the agency next door.
That's a fundamentally different conversation, and it requires a fundamentally different kind of person to lead it.
Capability is the bigger shift
I was talking to someone recently who runs an agency, and he told me he'd never really thought about AI beyond the efficiency angle until he saw what was actually being built with it. Things that his team wouldn't have known where to start with six months ago, projects they never would have even pitched because they seemed impossible for their budget or timeline, were suddenly becoming "we could ship this in a month" real.
That shift from "how do we do what we already do faster" to "what can we do now that we literally couldn't before" is where the magic is. But it requires someone who can hold both the conceptual thinking and the technical execution in their head at the same time. You have to be the person who can imagine the interactive installation or the marketing pipeline or whatever else doesn't exist yet, and then also be the person who sits down and builds the first version of it.
The vibe coder with an outside perspective
What makes this role different from just hiring a senior developer or promoting your most AI-curious employee is that the best version of this person brings an outside perspective. They're not embedded in the existing team dynamics, where everyone's been doing the same thing for years and the culture has calcified around certain assumptions about what's possible and what counts as "our kind of work."
Someone who comes in from outside, who's been building things across different contexts and has seen what's possible because they've been doing it for themselves and for other people, carries a different kind of energy into the room. They're not threatening the way a consultant is, because they're not there to evaluate anyone, and they're not a boss pushing mandates down either. They're just visibly doing the thing, building stuff that makes people go "wait, how did you do that?" That energy is contagious in a way no internal initiative or training program can replicate.
I think of it like how artist residencies work. You bring someone into an environment, give them space to do their thing, and the creative cross-pollination happens organically. Except instead of painting, it's someone who can look at your stalled marketing pipeline and say "what if we just build an API for that," and then have it working in an afternoon.
Why agencies specifically
Agencies already have the bones for this. They're multidisciplinary by nature, used to thinking in sprints, and they already work across very different client contexts in any given week. The structural problem is that the tools and processes haven't really changed in a decade while the capabilities available to those teams have exploded.
The agency that figures out how to bring this kind of person into their orbit, not necessarily as a full-time employee in the traditional sense but as someone who's embedded enough to understand the business and independent enough to not get pulled into the status quo, is going to have a massive advantage. They'll be pitching work their competitors can't even conceive of, and delivering it at a speed that makes the old model look absurd.
The role doesn't have a name yet, and that's fine
Maybe it's an "agentic operator" or just "that person who builds stuff and makes everyone else realize what's possible." The title doesn't matter. What matters is recognizing that the gap exists between knowing you need to change and actually changing, and that the fastest way to close it isn't a software subscription or a training program. It's a person who embodies what the future of the work looks like, sitting right there in the building, doing it where everyone can see.
The agencies that get out in front of this, the ones who bring this person in before there's a name for the role, are the ones their competitors are going to be trying to reverse-engineer two years from now.