Adopting AI is the easy part. Absorbing it is the work. 

The mistake most enterprises are making with AI is structural, not tactical.

They are treating it as a technology adoption — something to roll out, train people on, plug in. AI is not that. It is a business disruption causing huge shifts in operating models. It rewires how decisions get made, how data moves, how risk is held, where value accrues, what skills matter, who has authority. None of that is a software problem. It is structural.

This is the pattern I keep seeing in boardrooms across industries. The leaders who can read it — who recognize that what looks like a technology bet is actually a structural one — are making different decisions about where to invest, who to hire, how to organize. The leaders who cannot read it are pouring capital into pilots that will never be absorbed into the business. That is not a forecasting failure. It is a category error.

We started OnStak from the desire to help our clients solve the pain points. We have always worked with a deep belief that strong relationships and collaboration are what enable great results. Our work, which began in Application and Data Modernization, has enabled us to build a portfolio that makes the structural shift our customers are facing today possible — the architecture underneath, the governance built in, the integration with the rest of the business.

For almost fifteen years we have been migrating the unmovable, modernizing the untouchable and fixing the applications and data everyone gave up on. It put us in the position to be able to support our customers through the huge disruption that AI is causing to their operating models.

What we now need is to scale how we get to the companies that need it most.

I look for three things in a leader at this level: technical credibility with engineers, commercial instinct with buyers, and the stamina to do both at enterprise pace. Most candidates have one. Some have two. Fabio has all three.

But the credential that matters most for this moment is not in his title.

Fabio started his career writing software for air traffic control — a system where the cost of being wrong is not measured in revenue. He spent the next twenty-five years at Cisco, joining as a Consulting Engineer in 2000 and working through senior product, marketing and partnership roles across the United States and EMEA. Most recently he was Vice President of Engineering Alliances and Vice President of Customer Solutions Marketing.

The fact that he chose to come and build with us tells you something about where he thinks the market is going.

His mandate is straightforward to write down and harder to do.

Take what we have built to customers globally. Establish OnStak as the company that helps the enterprise absorb AI rather than just adopt it. Deepen the partnerships we are building with Cisco, NVIDIA and the leading neocloud providers. And build the product and marketing engine that matches the engineering work already in the ground.

I asked him to put his own job in his own words before he started. Here is what he sent back:

That is the right ambition, in the right voice, for what comes next.

My finance background taught me to watch for the things that compound.

Most of the public conversation about enterprise AI is still about flow — which model is best, which lab is winning, which benchmark moved this week. Flow is what gets covered. It is not where the value is decided. The value is decided by which enterprises actually rewire themselves around AI — the architecture underneath, the governance built in, the operating model redesigned around what AI now makes possible. That work compounds. The flow does not.

Five years from now, the gap between the enterprises that did the structural work and the ones that kept treating AI as a technology rollout will be obvious. They will have spent the same money. One cohort will have a business that compounds. The other will have a thicker layer of pilots and the same balance sheet exposure they had three years earlier. There is no middle outcome.