I worked with GecoSistema during my master thesis. My thesis tackled hydrological level forecasting with ML in Italian rivers. This meant I had a good idea of the scientific landscape around hydrological forecasting algorithms and the state-of-the-art.

After the Emilia-Romagna floods in 2023, GecoSistema partnered with the European Union Copernicus project. I was involved with the project because of my master thesis, and I was responsible for building a scalable system for hydrological forecasting in european rivers. This project will give european citizens earlier warnings about flash floods, making evacuation more effective.

The project was done with PyTorch and Lightning. It required me to implement a new state-of-the-art ML model from scratch. After this, I trained the model, tuned it and deployed it as an API on a Kubernetes cluster. Much more was involved in the project, and many things can still be improved about the system. However, the project was quite short (June/July were preparatory, August/September were for execution).

Awaves was a startup that I co-founded with the mission of building an AI DJ. This DJ is able to create mashups and DJ sets automatically. I was the technical mind behind the startup and in true startup fashion, that means I had to do a lot of things:

• Choosing the technologies to be used for development
• Designing the software architecture
• Writing a lot of code
• Handling deployment, service uptime, telemetry, etc.

The list above contains the very basics, and obviously there was a lot more to it. However, I would also like to specifically outline some major responsibilities with regards to the startup being an AI startup:

• Coming up with data strategies
• Setting up data pipelines
• Researching and creating ML models
• Training, evaluating and optimizing ML models

At Esports Team Twente, I was responsible for setting up the data pipeline for one of their teams' games. In addition to this, I did some projects with ML to give the teams insight in their communication and pathing efficiencies.

In the board of Hive01, I was responsible for organising lectures and events for students to learn the basics of entrepreneurship. I communicated with participants of the programme about what their needs were in the programme, and gave them feedback on their business models. I also communicated with external speakers and organised some workshops myself.

First year: University of Twente (Enschede, Netherlands)

Second year: KTH Royal Institute of Technology (Stockholm, Sweden)

The Processes of Change track teaches you to understand change and to promote it in practice. The first few courses were creating theory and understanding of businesses and change. After this, I did a few consultancy projects for the university.

In addition to this, the course had a focus on entrepreneurship and presentation skills. I was personally coached to improve my presentation skills and engage people with my communication.

My bachelor thesis (Finding “The Drop”: Recognizing the climax in electronic music using classification models) was about using ML to find the chorus in electronic dance music. It was recognized with a grading of 9.5/10.

• 3 years of academic and professional experience with building state-of-the-art machine learning systems.
• Good understanding of - and experience with - end-to-end machine learning projects (data exploration, feature engineering, training, tuning, deploying).
• Experience working with stakeholders, both internally (designers, operations, business, etc.) and externally (clients, users, project financiers, etc.).