Modelling is, in general, a fundamental methodology to understand possibly complex systems, and study their behavior as a function of key parameters of the system itself, and of the environment where the system is embedded. As such, having familiarity with modelling techniques is an important skill for decision making. Models are essentially abstractions of a real system. In order to be useful, they need to be simple. This means that models inevitably abstract features of the real system that are considered not fundamental, to simplify them. This is however a double-edged sword. Understanding the assumptions and approximations behind a model is fundamental to be aware of the models limits, and thus understand which outcomes of the models can be considered to accurately describe the system’s properties, and what could generate overinterpretations. Machine learning can be seen as a way to extract models from data. Due to the fundamental importance of BigData and AI in the coming years, machine learning techniques are gaining more and more importance. As any model, it is fundamental to know which the limits and assumptions are, used to generate them. Often, there is a perception that Machine Learning is a sort of “magic black box” that can be used to extract knowledge from otherwise incomprehensible huge amounts of data. While this is certainly the case in many times, it is also true that Machine Learning tools need to be mastered and properly used, otherwise they could provide, as any modelling tools, inaccurate or plainly wrong indications. The aim of the module is thus to walk attendees through some key modelling and machine learning techniques. As a comprehensive presentation of modelling tools and Machine Learning algorithms is impossible to be provided in a single module, the main goal would be to provide some concrete examples of wrong approaches to modelling and Machine Learning, so as to provide some cautionary perspective on the use of results obtained from these tools.