Taking a Closer Look at the Scuderia Ferrari and Palantir Partnership

Saahil Barai
5 min readJul 26, 2021
Photo by Spencer Davis on Unsplash

Palantir has long been a company that I have admired for the work they do in solving some of the hardest problems that are soon to be and currently faced by businesses. The software Palantir is creating is an end-to-end framework that allows a business to integrate data from multiple sources and intelligently query the data to bring actionable insights to the forefront of a business’ decision-making process. Consequently, as growing amounts of data are collected, Palantir’s software is going to be transformational for businesses in the years to come. Palantir’s partnership with Scuderia Ferrari is early evidence of their potential to change the way we interact with data.

Over the past couple of months, I have been watching Formula 1: Drive to Survive, a Netflix documentary series that offers a behind-the-scenes look at the teams, drivers, and races of the Formula 1 World Championship. As a fan of technology, I was amazed to learn of the vast amounts of data collected on the car-driver pair. The show provides an incredible look at the data and analytics that goes into the sport week in week out. So, when I learned of the Scuderia Ferrari and Palantir partnership, I was thrilled. Mulling over the news, I had many aha moments that got me excited both for the future of Formula 1 and for the future of Palantir.

Formula 1 teams live and die by time. Time is a team’s only creed.

The most evident examples we see of this are on race day and qualifications. On top of being a determining factor to assess a driver’s performance and placement, the team relies on real-time data to make informed decisions during races.

Each Formula 1 car is equipped with an electronic control unit (ECU). The ECU acts as a data collection unit as well as a data transmission unit amongst many of its other processes. It is responsible for taking data from over 200 sensors and relaying the data points back to the team in real time.

With each of these 200 plus sensors rapidly capturing data and the ECU relaying back the data to the team, it becomes apparent that the team is receiving an extremely large amount of data during each race. Making sense of all this data in real time then becomes a tremendous challenge. While a subset of sensors can allow the team to make cursory analysis, there remains a vast amount of untapped potential in the data. For example, the team can look at the temperature sensors for the brakes and if overheating they can alert the driver to let off the brakes. However, if we want to ask the question when is the most appropriate time to pit, quantifying an answer becomes difficult. This now involves looking at many more streams of data, identifying which streams hold importance to the question and running them through a predictive model. Palantir enables exactly this. It allows teams to tap into the full potential of the data in real time and go beyond a cursory analysis.

The time constraints that that are less visible to us take place in-between races. The fast-paced nature of the Formula 1, with races happening biweekly and sometimes even weekly, provide only so much time for teams to analyze race data, make improvements to their vehicle and provide feedback to the driver. Inefficient exploratory analysis of the data can dig into this valuable iteration time. As such, it is important that the data organization is efficient and lends itself to the business’ needs.

Palantir’s dynamic ontology allows for structuring the problem space and data to fit the Ferrari’s needs. There is a great video put out on this concept by Palantir.

Simply put, an ontology seeks to classify and categorize entities. The ontology within Palantir’s software is modeled using objects, properties, and relationships. In this instance, Ferrari could break down their car down into parts, part properties, or relationships between parts or systems. When interacting with the software, the ontology layer and its labels will stand in front of the data. The data is abstracted behind the physical part. Instead of being met with data sprawling over many rows and columns, the user is met with an organization that is a reflection the real-world car.

The beauty of this is twofold. Not only will this enforce a software-wide common language, but it will also enforce a business-wide common language. All the individual users interacting with the software will then be subject to the same terminology, spurring ambiguity free collaboration.

Another, importance of this ontological structure is that it allows flexibility in the way data is interpreted and modeled. The structure is dynamic. Ferrari can make their ontology as specific or as general as they like. For instance, we can model the brake system of a Formula 1 car with lots of detail as follows:

A more specific ontology of a brake system consisting of many objects that have properties.

Or more simply as follows:

A more general ontology of a brake system consisting of only properties.

By allowing the team to speak the same language that is tied to the physical world, eliminating the need for data hunting, and enabling organizational flexibility, dynamic ontology grants Ferrari with the time to focus on what is most important.

Where this organization becomes even more crucial is when we take a look at some of the rules and regulation changes taking place for the upcoming 2021 season.

  • A cost cap that limits each team from spending above a certain threshold on general operations. This cost cap is set to reduce yearly.
  • A sliding scale for aerodynamic testing that limits the top finishers testing time in the wind tunnel or simulations and awards more time to the lower finishers.
  • Reduction of practice session timings that will give the team less time to tweak their cars prior to qualifications and racing.

Each of these rules serve to constrain teams and even the playing field. Teams will have to look to new methods to differentiate themselves and gain an edge on race day. Fortunately for Ferrari, they can look to Palantir for this edge.

So, what does all of this mean for Palantir outside of this partnership?

It shows that Palantir’s software can be relied upon in a time-critical, rigorous setting to deliver valuable insights. Moreover, the software performing in the fast-paced world of Formula One acts as a beacon to other businesses across industries who are also constrained by time and simultaneously desire peak efficiency. With such a significant addressable market to conquer, Palantir’s brightest days are ahead.

--

--