Surprisingly Useful AI Article Enhancements
Formula One’s newest team has revealed how it leaned on enterprise software vendor IFS as a foundational pillar of its rapid build-up from a standing start, and is now using the company’s AI-powered Intelligence platform to tackle data quality problems that would take human staff weeks to resolve manually.
Chief Information Officer of Cadillac F1, Chris Green, appeared on stage at the IFS Unleashed conference to detail the racing team’s journey from a blank sheet of paper to a competing constructor in under four years, with IFS embedded from the very beginning.
“We chose IFS before we bought our first coffee machine,” Green told the audience.

Getting permission to race
Green explained that Cadillac F1 has spent roughly four years assembling the organisation and is currently three-quarters of the way through its debut racing season. But before any car could turn a wheel in competition, the team first had to clear a significant regulatory hurdle.
“The bar to enter Formula One is very high, and rightly so,” Green noted. “You really have to demonstrate your abilities and your capabilities. And part of that is not just being able to design a racing car. You need a great business case to actually show you’re ready to do it.”
IFS was central to making that case. The team began implementing the platform two and a half years ago to underpin its core manufacturing and financial processes. It took a further six months after that initial deployment before the team received the green light to compete.
Once approval came, the clock started ticking. Green described a 366-day sprint to prepare for the season-opening race in Melbourne.
Scaling at an extreme pace
The speed at which Cadillac F1 has grown placed enormous demands on its systems. Green revealed that during the team’s most intense recruitment period, 147 people joined in a single week, representing a quarter of the entire workforce at that point.
Many of those new hires arrived to lead departments that did not yet exist. The ability to onboard them directly into IFS from day one proved critical.
“They were able to come in and start to use IFS from day one to ensure that we were ready to race,” Green explained.
He described IFS as providing a “drumbeat” for the organisation during its build phase, with the system serving as the backbone against which milestones were set and knocked down. The team deployed its procurement and payment processes within approximately four weeks of standing up the cloud environment.
“That milestone, putting IFS in place, got us in really good stead as we hit our next milestones,” Green reflected. “Getting the car prepared, getting a car to the first winter tests, getting the car to Melbourne.”
Early access to IFS Intelligence
IFS Global Chief Technology Officer, Vaibs Kumar, introduced Green as one of a small group of flagship customers given early access to IFS Intelligence, the company’s new AI-driven platform that allows users to interact with their business data through natural language.
Green disclosed that the team had been using Intelligence for roughly five weeks at the time of his appearance, though he personally had only been granted access for two of those weeks.
Related: Best Business Laptops for work & school
Related: Best Gaming Laptops
Related: Best Portable Laptop
His initial use has been practical rather than experimental. As CIO, Green has been using the platform to interrogate financial data for budgeting and planning during the team’s current budget cycle.
“I’ve been using it to access the data in the system to help me understand my cost centre, to help me understand what I need to be planning for, to help me in the budgeting cycle that we’re deep in,” he explained. “Previously I would have to ask my teams to build the reports or dashboards. I can just get in there and ask questions with the data straight away. That’s been really powerful.”
Hunting data anomalies at speed
The more consequential use case, however, involves identifying and correcting data quality issues that have accumulated during the team’s rapid build-up. Green was candid about the fact that building a racing team at speed inevitably produces imperfect decisions.
“Part of building out a team really fast is that you have to make quick decisions,” he acknowledged. “Some of those decisions may not have been as good in hindsight as we thought.”
The specific problem Green described centres on the serialisation of manufacturing parts. During the team’s early growth phase, certain components that were intended to be serialised were not. In other cases, the team later realised that parts needed to be serialised retrospectively. The implications ripple across multiple business functions.
“Serialisation is used in our financial reporting, it’s used in our engineering processes, and in our safety processes for running a safe car,” Green explained.
Compounding the difficulty is the fact that many of these parts have since been consumed into assemblies, which feed into sub-assemblies, which build into larger structures within the car. The unit the team handles at the racetrack is not the individual part but pre-assembled sub-sections. Correcting the data at its source requires tracing transactions that sit 20 to 50 layers deep within the system’s data structure.
“To do this manually, you have to effectively look at the history of the entire part and reverse it all back out in the system, correct the data, and then replay all the transactions back in,” Green described.
He quantified the scale of the problem: correcting a single part manually had taken the team two weeks.
From two weeks to two minutes
Green expressed confidence that IFS Intelligence can compress that timeline dramatically. The platform’s understanding of IFS’s transactional data model and the history of individual parts allows it to automate the reversal, correction and replay process.
“We believe we’ll be able to use Intelligence to reverse those transactions for us, go back, reverse everything out, correct the data, and then play all the transactions back in,” he outlined.
The expected reduction is stark: from two weeks per part down to two to five minutes.
Green added that the team had received a fresh deployment of code on the morning of his stage appearance and that staff were testing it live during the conference.
“We deployed a new set of code to us this morning and the team is testing it live at the moment,” he confirmed.
Beyond correcting known issues, Intelligence is also helping the team discover problems it did not know it had. The rules used to determine how parts are classified are being applied across the system to flag additional items that may have been categorised incorrectly.
“It’s helping us spot where we’ve got more corrections to make,” Green added.
Natural language for occasional users
Green also outlined plans to use IFS Intelligence to improve how less-frequent users interact with the system. While Cadillac F1 has a core group of staff who use IFS daily, a significant number of employees only dip into the platform occasionally, and those users are more likely to introduce data quality issues.
“One of the challenges that we see is making sure those people who are dipping into our system can actually interact with it in a way that allows us to have high data quality,” he explained. “If we get the data correct at the beginning, everything downstream flows. If we get it wrong at the beginning, the cost continues to build up.”
Green revealed that the idea for this application traces back to the previous year’s Unleashed conference. A colleague named Laura was sitting in the audience when IFS demonstrated its copilot capabilities on screen.
“She was tapping me on the shoulder and she said, I want that,” Green recounted.
The plan is to build a chat-style natural language interface for the requisitioning process, allowing team members who are not regular IFS users to order parts and services without needing to navigate the system’s standard screens.
“We basically want to roll this out so that the team can interact with requisitioning of parts and goods and services through IFS, but doing it in a natural language way for those that are not using the system day to day,” he explained.
Green characterised the capability as eliminating what would otherwise have been a separate IT project to build a custom front end.
“So that’s one less IT project for me,” he concluded.
What comes next
Green indicated that the team’s CEO and one of its race drivers would be appearing at the conference later in the week to discuss the broader Cadillac F1 journey.
IFS Global CTO, Vaibs Kumar, framed the Cadillac F1 deployment as an example of the co-work mode within IFS Intelligence, where a human user leads and the AI assists. He positioned it alongside demonstrations of the platform handling tasks from shift handover preparation to working capital analysis across multiple legal entities.
For Cadillac F1, the immediate priority is to fix the data that was laid down in haste during the team’s rapid formation, arm occasional users with a more forgiving interface and free up technical staff from manual data correction work that consumes weeks of effort for outcomes that AI can deliver in minutes.
The broader signal from the team’s experience is that even organisations built from scratch in the modern era, with no legacy systems to contend with, still accumulate data debt at speed when growth outpaces process maturity. The question, as Green framed it, is whether AI tools can close that gap before the cost compounds further.
Last Updated on October 16, 2026 by Nick Ross



