How the New York Jets Are Advancing the Ball for an AI Forward Future with Monte Carlo
AUTHOR | Virna Sekuj
Background
The New York Jets’ ownership, led by Chairman Woody Johnson, set a mandate to build “the best AI decision-support infrastructure in all of sports.” To lead it, the Jets created an innovative and entirely new role: Iwao Fusillo, hired in 2026 as Chief Data and Analytics Officer, with a remit spanning both football analytics and business analytics, plus application development and AI strategy, all reporting into one office. Fusillo previously built data and analytics functions in professional sports, at General Motors, and at PepsiCo — where he first worked with Monte Carlo, making the Jets his second engagement with the company.
Challenge
When Fusillo joined the Jets roughly eight weeks before the 2026 NFL draft, meaning the data foundation and the AI program had to be built in parallel rather than in sequence — with no existing playbook to draw from.
The stakes were high from day one: AI-generated decision support was going in front of the general manager during the draft and informing multi-million-dollar sponsorship renewal decisions, where, in Fusillo’s words, “it has to be correct, and there can’t be any caveats.”
As the Jets’ AI footprint scaled from 7 agent deployments to 51 in about three months, the manual review model that worked at the start stopped scaling: Fusillo could personally review every AI deployment at 7; by 51 it required him and his direct reports. It would be impossible to do this at a scale of 100, 200, or 300 deployments.
Quality problems surfaced as the fleet grew, and they didn’t announce themselves — a data pipeline would silently go unrefreshed, a schema would quietly break, or an agent would fabricate an answer rather than flag uncertainty.
One concrete flashpoint: about three weeks before the team’s pre-draft scouting combine, Fusillo discovered that the team’s head of medical and his staff were on pace to spend roughly four weeks manually analyzing unstructured medical dictation transcripts generated at the combine. This amounted to thousands of pages of content from hundreds of players — a timeline that would have run well past the window they actually had to work with.
Where Monte Carlo Fit In
The Jets onboarded Monte Carlo in 2024, and it was already an established part of their business data foundation when Fusillo joined.
Fusillo has explicitly reframed data engineering at the Jets internally as “a trust layer that underlies every part of our AI program,” treating data reliability and agent behavior as the same problem rather than two separate ones, since “the agents will amplify whatever data quality you feed them, and the data quality will amplify the effectiveness of the agent.”
On top of that foundation, every one of the Jets’ AI deployments is required to have a named owner, defined performance metrics, and an explicit autonomy level (human-in-the-loop vs. full autonomy). Four leaders across data science, software engineering, and business analysis were personally certifying every AI deployment.
Recognizing that model won’t survive the next order of magnitude of growth, the Jets and Monte Carlo are prototyping what Fusillo calls an AI-deployment registry: shared visibility across use cases, a way to surface and reuse top-performing deployments, and some degree of self-certification. Fusillo is candid that “model registries and self-certification… [are] commonplace at a tech company, it is not at a sports team,” so this joint work represents a significant innovation for the industry.
“The agents will amplify whatever data quality you feed them, and the data quality will amplify the effectiveness of the agent.”
Iwao Fusillo, Chief Data & Analytics Officer, New York Jets
Day to day, new AI use cases at the Jets are required to start and remain as a supervised prompt for days, weeks, or in rare cases months before they’re allowed to become an autonomous agent — surfacing data-quality issues or output drift while a human is still watching, before any autonomy is granted. High-visibility outputs keep a human in the loop even after deployment: the Jets’ daily executive media-intelligence agent, for example, is fully autonomous end to end, but the communications team reviews its output before it reaches the executive team, and the team also runs a second and third LLM as cross-checks to compound down the error rate.
Result
With that foundation in place, the Jets’ AI deployment count grew from 7 at the end of Q1 to 51 roughly three months later — more than a 7x increase — spanning every major business and football function.
97% of the Jets’ front office is now active on Copilot. The AI-powered sales coach reviews hundreds of thousands of ticket sales calls a year against a rubric designed by the Jets’ own sales managers, delivering team-wide Monday-morning insights and rep-level coaching at a scale Fusillo says “no human manager could ever possibly deliver.”
Plus, the agent built ahead of the team’s pre-draft scouting combine took a four-week manual review of unstructured medical dictation transcripts down to two days. Software engineering and data science teams operating under the Jets’ “hybrid workforce” model are seeing 2-3x, in some cases 4x, productivity gains today, with the team aiming toward 20-30x.
“We can’t rely on hope. Hope is not a particularly good strategy. Partnering with your company and your team is a great part of the strategy, because it gives us the types of tools we need to understand the underlying data, the behavior of our models, and the quality of our outputs.”
Iwao Fusillo, Chief Data & Analytics Officer, New York Jets
Watch the full conversation below.
How the New York Jets Are Advancing the Ball for an AI Forward Future with Monte Carlo
New data lineage and impact analysis capabilities help data teams understand incident severity and proactively communicate to users consuming data via Hex.
