Reflections from the Executive Data & AI Leadership Dinner
Kembara, JW Marriott Phoenix. More than forty leaders, one long table, and no agenda to sit through.
An evening built around talking
We set one rule for the night: nobody presents. No decks, no panel, no product walk-through. Guests arrived onto the lawn as the light went down over the desert, and the first conversations started well before anyone sat down to eat.
What followed was the kind of exchange that is almost impossible to schedule. Leaders compared notes on the programs that stalled, the migrations that took longer than promised, and the governance work nobody budgets for until it bites. People asked each other questions instead of trading credentials.
Who was in the room
More than forty senior leaders from banking, payments, insurance, retail and enterprise technology teams across the Valley. Heads of data, analytics and engineering sat alongside architects and business owners, which is exactly why the conversation stayed practical.
The seating was deliberately mixed, so most people spent the evening next to someone solving the same problem in a different industry. By dessert, several of them were still at the table comparing approaches.
What the room kept coming back to
Five themes that surfaced again and again across the evening
A business-led roadmap
AI work lands when it is tied to objectives the business can measure and name.
An AI-ready foundation
Governance, quality and access are the unglamorous groundwork everything else sits on.
Teams that work together
Business, technology and data functions pulling in one direction beats any single tool.
Value in increments
Ship something useful this quarter rather than waiting on a multi-year program.
Federated by default
Query data where it already lives instead of moving it for the sake of moving it.
Our guest for the evening
Sanjeev Mohan
Principal, SanjMo · Former VP Research, Gartner
Sanjeev spent his time on where data and AI architectures are actually heading, drawing on the research behind his new book rather than on anything vendors are currently promising. His argument was that the centralised model many enterprises spent the last decade building is now the thing slowing them down.
The alternative he described is federated: let teams query data where it already sits, put governance at the point of access rather than at the point of copy, and stop paying the cost and risk of moving everything into one place before any value appears. Several people in the room said afterwards it reframed how they were thinking about their 2027 plans.
Voices from around the table
On building the foundation first
VP Engineering · Financial Services
Shared hard-won experience on building AI-ready data foundations: strengthening governance without slowing delivery, and getting business, engineering and data teams working together early enough that time-to-value stops being a negotiation.
On a roadmap the business recognises
Head of Core Banking Technology
Made the case for a clear, business-led roadmap and modern architecture that widens access to trusted data, cuts operational complexity, and gives teams room to innovate at scale instead of queuing for it.
On adoption inside a regulated enterprise
VP, Global Data Intelligence Platform · Financial Services
Talked through the opportunities and the realities of AI adoption in a large, highly regulated enterprise, where every initiative has to show measurable business value while navigating complexity and compliance.
Ashish Baghel
Founder & CEO, NucleusTeq and NuoData · Host
Ashish opened the evening by explaining why there would be no presentation. The most useful thing any of us could do, he said, was compare notes honestly about what has worked and what has not. He closed on the same idea: the organisations getting value from AI are rarely the ones with the most ambitious strategy deck. They are the ones that shortened the distance between a question and a trustworthy answer, and then did it again next quarter.
We missed you at the table
This room was put together carefully, and your name was on the list for a reason. The conversations were candid and specific, and we would rather not have you read about them second hand.
If you have fifteen minutes, we would like to share what came out of it and hear what is on your roadmap for the year ahead. No deck, same as the dinner.
NucleusTeq | NuoData