Could you describe your activities?
At Flit Sport we design, develop and market an online app that offers personalised training plans for runners using artificial intelligence. We take account of the athlete's objective, physical abilities and availability to create customised plans.
What are the technological challenges facing the sector? And what’s next for the start-up?
The biggest technological challenge lies in defining each runner’s profile accurately. Training data is readily available. Today, a large majority of runners use apps to track their runs and they often have smart watches or heart rate monitors. These tools help us understand them better.
When we can access this data, the challenge is being able to analyse it properly and interpret the rest. There is a big gap between what we are capable of measuring and what really goes on at a physiological level (tiredness linked to non-measurable factors, different individual profiles) and we still have relatively little understanding of the physiological processes linked to training. There is currently a lot of research on the subject, but there is still a long way to go.
The next phase of the project is to launch the app among beta testers (August 2021) which will be followed by a round of fundraising with business angels to accelerate the subsequent marketing phases (late 2021).
What did you do before setting up the company? What led to this start-up project? What support did you receive from Inria?
Before setting up the company, I studied at an engineering school (Mines Paristech) before doing a CIFRE PhD (University of Bordeaux - CATIE) in sports science at the STAPS in Bordeaux. I was supported by the student incubator of the University of Bordeaux (UBeeLab) for a year and then joined Unitec through the SporTech incubator and Inria Startup Studio in parallel. Aurélien worked in several start-ups with his Master’s in computer science before carrying out a PhD in the POTIOC project team at the Bordeaux-Sud-Ouest Inria centre. His PhD focused on classifying neurophysiological states through machine learning. As a top-level triathlete, it was only natural that he should be interested in the subject!
The project evolved into its current state gradually, but the basic idea mainly came from our research, knowledge and shared interest in the world of sport. After deciding we wanted to launch a business in the field of sport and data analysis, we identified the need through discussions with numerous sportsmen and women with different profiles and backgrounds.
Inria’s support has been key at different levels. In addition to the scientific aspect of supervising Aurélien's PhD, the Startup Studio put us in contact with research engineers who worked on the technical aspect, besides covering my salary and providing invaluable support by offering advice on all matters connected with the business side of the project.