For Zapier's Head of AI Transformation, Hiring Is The First Lever For Successful Adoption

When companies talk about getting good at AI, the conversation usually starts with tools, training, and rollout plans for the people already inside the building. That skips the first lever. Before any of that pays off, an organization has to stop hiring people who drag the average back down, which means the resume and the polished interview matter far less than what a candidate can actually build and how they think about the work.

Tackling this challenge is Dan Slagen, Chief Marketing Officer and Chief AI Transformation Officer at Zapier, where he leads the company's AI strategy alongside its global marketing organization. A Forbes Top 50 Entrepreneurial CMO and an AI-transformation instructor, Slagen has spent the past year rebuilding how his teams work and, upstream of that, who gets in the door.

"First and foremost, any and all hiring has to have an AI fluency bar," he says. In his experience, this means getting specific about where the bar is and how you read it in a room.

Hiring can make or break an AI strategy

Slagen frames talent as the leading edge of everything else a company is trying to do with AI. Ask him how an organization actually gets great at it, and he starts before onboarding, before tooling, before the measurement dashboards he obsesses over. "The first thing is, you have to stop new people from coming in the door that aren't at the level that you want," he says. The mechanism is a standard he applies to every role. "We look at where people are in their own AI journey. There are a lot of ways to go about it, but I literally just ask people."

The open-ended question does more work than one might expect. "You can start to calibrate based on the answers that come in where someone's at with AI," he says. From there, the questions sharpen: What's the biggest system you've implemented with AI? How are you training your own team? Which projects are you converting from manual to automated? The point isn't to tick boxes on a checklist. Rather, it's to clear a bar the hiring team feels good about defending.

Three levels of AI fluency

Based on this analysis, Slagen sorts candidates into three levels, and the language to describe them is deliberately plain. "We look at capable, adoptive, and transformative," he says. "Capable is the minimum bar for anyone right now. Adoptive is ahead of the curve, and transformative is someone who's pushing the limits of what we see."

The obvious move is to chase the transformative candidate, but Slagen thinks that instinct is mostly a trap. The supply is thin and the people at that level rarely want to move. "What percentage of people are absolutely perfect at AI and want to leave the job they're at to go somewhere else? You have to try to find the best player on the worst team. Most people at an awesome company who happen to be in the fortunate position of killing with AI right now probably are not going to want to leave."

The candidate Slagen actively hunts for is the one sitting just below the ceiling, climbing towards it. "What I really try to look for are the people in that adoptive stage who are right at the cusp of their hockey stick growth," he shares. "If you can just give them one thing that unlocks it for them, they'll spike like crazy."

He describes a recent hire that tested the theory. "We were loving this person in the interview but a little worried about the AI part. They weren't quite at the transformative level, but we felt like this was someone worth taking a risk on." The unlock was small and specific. "The thing that unlocked it for him was creating a digital twin, almost like an AI buddy, that could help him do all the things that he wanted to do. Once we gave him that, he has scaled out so many AI systems across the company."

This is the signal Slagen wants recruiters tuned to. "That's the diamond in the rough that you're trying to look for."

What survives when the skills look identical

Slagen likes assessments, build exercises, and technical evaluations, but he thinks the ground under all of them is shifting, and it changes what the interview is really measuring. "I tend to think where this is all going is everyone has the exact same level of skill, in a way," he says. "For me, it comes down to vision, curiosity, thinking about things differently, and rebuilding versus just improving. Who has the ability to think that way more than anything else?"

He sees the distinction show up at every level, including junior ones. "You can see the people that are building with vision versus just building because they were told to build something. They can see a bigger connection."

It's also how he screens for authenticity in an era when a candidate can look up a polished answer in seconds. "You start to calibrate what sounds like a different answer, what doesn't sound like you just looked it up with a prompt." The questions he returns to are the ones a rehearsed candidate can't fake for long: What systems have you built? What are you most proud of? Where is the balance between human and agentic work? How do you think about the future?

None of this is a filter you can run automatically, and that's precisely Slagen's point. When the paper credential and the tidy interview response stop separating candidates, the thing left to assess is how a person actually thinks and what they've genuinely built. The interview must get better at reading those signals, because they're the ones left worth trusting.