👋 Hi, I’m Andrea, a self-employed engineer who creates mathematical models of sport physiology and performance.

I love to work on:

🫁 AI applied to cardiopulmonary exercise test data

🚴 Cornering strategies in road cycling

🤖 AI training prescription & assessment

📈 Continuous Glucose Monitoring (CGM) data

🔍 WHY

I’m equally interested in each phase of modeling: conceptualization and formalization, data collection and processing, model development, and model deployment. I always value models that can teach us more in return. I love adding intelligence to the process. Ironically, more intelligence usually means less technology. And that’s the way it should be: sensors come and go, but learning is forever.

Modeling lies at the heart of how we understand the world and its emerging properties. Mathematical modeling, in particular, describes a system using mathematical terms. The clear advantage of having our models written in a communicable language is that we can share our ideas with the world. If we’re lucky, our ideas push human knowledge forward.

I enrolled in an engineering course to study robotics and ended up studying the most wonderfully made machine: the human body. Sport performance requires a combination of the best qualities of the human body, making it human movement in its best form. Physiology is what makes movement possible, hence human movement in its fundamental terms.

🛠️ HOW

I apply a blend of methodologies to the development and deployment of mathematical models. I collect or retrieve experimental data, then build a mathematical model that can best describe their behavior. I either follow first principles, writing differential equations by hand, or I blend mathematical equations with hard numerical approaches, such as neural networks or machine learning models.

I love to deploy my models on basic web applications with Flask or carry out real-time projects with Tornado and Raspberry Pis or Arduino boards. I enjoy creating POCs so stakeholders can try and break my ideas, as well as attempting something at the edge of the unknown. I deploy on AWS or Heroku, mostly creating Dockerized versions of them. I create endpoints and API services for them, and I might want to sell their usage one day.

I have been making heavy use of Python for 8 years now, but I come from a MATLAB background—which is not always a point of pride. I challenge myself by trying JS scripts and managing to print endless pages of warnings, busy servers, Node package errors, and exceeding memory limits on Heroku. I then engineer solutions to get rid of those… and create new ones, even more challenging.

I have a deep admiration for those who think in statistical terms and probabilities.

In recent years, I have been profoundly changed by the vectorization of tokens and by the concept that the trajectories of ideas in the thinking space are deviated by information density as much as light is bent by mass in spacetime.

🚀 WHAT

In short: I apply a scientific research mindset to the startup and technology world of practice. I’m taking part in the growth of Athletica.ai, an AI-driven platform that follows the basic science principles of HIIT Science. I have been involved in Supersapiens within their Science Team. I also work for other startups currently developing cutting-edge technologies, but I’m not sure whether they would like me to mention their names yet.

In the pursuit of an impossible academic career, I love writing controversial research papers that take eight rejections to get published. I challenge the status quo with modeling ideas that require a shift in thinking about human body biomechanics and physiology.

I help the community by sitting on the editorial board of Sports Engineering, where I am a proud Associate Editor. I act as a reviewer for several top-50 journals in sports science, handling at least eight papers per year. I quite enjoy exchanging ideas with people in academia and with eager young minds.