S³ 2025

Rockets are the only machines powerful enough to reach orbit, carrying satellites, astronauts, and scientific instruments beyond our atmosphere. In 2024 alone, there were more than 180 rocket launches, highlighting the growing interest of both research and industry in reaching space. Placing spacecraft in orbit is essential not only for exploring the universe but also for studying our planet and learning how to protect it. I know rocket science might seem incredibly hard—and don’t get me wrong, it is—but are you curious about how rockets actually work? How do they generate enough thrust to defy gravity, and how can we control their flight so precisely?

In this project, we'll dive into the physics and engineering that make rocket launches possible. You will explore multiple branches of physics— dynamics, thermodynamics, and aerodynamics—while applying an engineer's mindset to solve real-world challenges. You will discover how rocket engines produce thrust, why stability is critical, and how each component influences performance. Through hands-on experiments, you will test how factors such as size, shape, and balance affect a rocket’s flight. Finally, we will evaluate how well engineering models describe reality by launching a small rocket and comparing the real-life results with mathematical predictions. By the end of this course, you'll gain not only a deeper understanding of rocket science but also insight into complex physics derivations and how engineers apply them to real-world problems.

Diabetes, known since ancient Egypt as the "sweet disease," was once treated with starvation — limiting patients to just 240 calories a day (roughly half the size of a Snickers bar). In 1921, insulin was discovered in dog pancreases, sparking a revolution in insulin production from pigs and cows. This continued until 1982, when the first recombinant human insulin was produced in E. coli, offering a purer, safer, and cheaper alternative.

In this project, we’ll follow the steps of insulin's journey. Just as insulin was produced in bacteria, we’ll introduce genes for fluorescent proteins into bacteria, make them produce the proteins, and purify them. Along the way, we’ll learn to use tools for visualizing DNA sequences and protein structures. This project will give you a hands-on understanding of how genetic engineering is used to create life-saving proteins like insulin, and how we can apply these techniques to other proteins. If you're ready to explore how biotechnology is changing the world, join me on this journey!

We live in a time where Machine Learning is getting more and more important, with AI models being used for an increasing number of tasks, such as sorting through job applications or identifying academic fraud. Therefore, it is important to make sure that computers make good decisions that do not unfairly discriminate anyone. For example, nobody wants their university application denied because of their race or their gender. But what exactly does it mean to make a fair decision? How can we teach that to an algorithm? And how do algorithms “decide” or “learn” something, anyway?

Those are the questions we will attempt to answer in this project! To that end, we will look at different definitions of fairness, how to express them so that a computer can understand them and evaluate both how accurate and how fair or unfair some machine learning models are. You will learn how to evaluate algorithms and we will discuss how different evaluation methods are in conflict with each other. We will also consider what kind of role datasets and potential data bias play for fairness in machine learning. And we will be working on how to make machine learning models more fair. You will learn the basic theory behind a few standard machine learning models and train and test models of your own. Our work will focus mainly on classification models – which sort data samples into pre-defined classes – and we will mainly use smaller models rather than deep neural networks. We will work with the programming language Python, but no previous experience in coding is required.

When we think of telescopes, we often picture glass lenses peering into the night sky. But there’s an entirely different kind of telescope that doesn’t rely on visible light at all—radio telescopes. Radio telescopes don’t collect light like your average optical ones. Instead, they collect radio waves—a type of electromagnetic radiation, just like visible light, but with much longer wavelengths. These radio waves are naturally emitted by various cosmic sources: stars, galaxies, nebulae, pulsars, black holes, and even leftover radiation from the Big Bang. Some of these waves have been traveling for billions of years before they reach Earth. You may think that to capture these radio waves you need to have large radio telescopes built by top experts in the field. But can we record the whispers of the cosmos by ourselves?

That is a question which we will try to answer. We will first dive into the world of atomic physics because we need to understand what happens when atoms produce radio waves. That is essential for our goal to capture them. After that, we will explore the world of radio frequency and radio communication, most importantly, the electronics behind all of that. The telescope’s large dish acts like an ear, collecting and focusing these signals onto a sensitive receiver, kind of like a giant satellite dish tuned to the universe. Once collected, these signals are converted into data graphs, images, and even sounds, using sophisticated computers. What looks like a simple “noise” is actually full of structure. Using signal processing we can determine: how fast a galaxy is moving, how fast our planet rotates around the sun, and around its own axis, or even spot strange, unexplained signals from deep space.

This year’s Summer School of Science will host a project developed in collaboration with the Department of Earth and Planetary Sciences at ETH Zürich. The project will explore cutting-edge topics in geophysics and climate change.

Full details will be announced soon. In the meantime, take a look at last year’s ETH Zürich collaboration project to get a sense of what's coming.