(Podcast) Welt der Werkstoffe - talk, Folge 24, Jakob Zeitler: Einsatz von KI in der Werkstoffforschung

Feb. 1, 2023

I had the chance to talk about Machine Learning in Material Science with Professor Bonnet on his show "Welt der Werkstoffe".

Professor Martin Bonnet runs a podcast on material science called „Welt der Werkstoffe - talk“. I joined him for an episode earlier this year to discuss current research of machine learning to material science. I really enjoyed the conversation, also because we were able to present a lot of the key cornerstones of Machine Learning for the audience and make an otherwise quite complicated topic easily accessible to his listeners. Check it out here, and make sure to use the YouTube auto-translate feature (see "CC") if you don't speak German:

Here is the podcast description, translated to English:

Prof. Bonnet talks to Jakob Zeitler in the new episode "World of Materials - talk". Jakob Zeitler graduated from high school in Germany, then studied in England and America and is currently doing his doctorate at University College London (UCL). He is writing his doctoral thesis on the subject of artificial intelligence for closed-loop automation of experiments in materials science. This episode is about how to integrate machine learning (also known as artificial intelligence or AI) into the laboratory and thus achieve added value for materials research. By collecting data and applying algorithms, one can create models of material behavior using many parameters. These models can then be used to derive insights for further experiments. Some examples are discussed below. Machine learning does not replace the researcher, but provides data and models that would take a long time and a lot of money to produce in real laboratory tests. Listen to the new episode if you want to know how the world of AI can be combined with the world of materials and what the laboratories of the future can look like.

Here is the original German episode description:

Prof. Bonnet unterhält sich in der neuen Folge „Welt der Werkstoffe - talk“ mit Jakob Zeitler. Jakob Zeitler hat in Deutschland sein Abitur absolviert, anschließend in England und Amerika studiert und promoviert zurzeit am University College London (UCL). Seine Doktorarbeit schreibt er über das Thema Künstliche Intelligenz für Closed Loop Automatisierung von Experimenten in der Materialwissenschaft. In dieser Folge geht es darum, wie man das Machine Learning (im Volksmund auch Künstliche Intelligenz oder KI genannt), ins Labor integrieren und damit einen Mehrwert für die Materialforschung erreichen kann. Durch das Sammeln von Daten und das Anwenden von Algorithmen kann man Modelle zum Materialverhalten erstellen und hierbei viele Parameter anwenden. Durch diese Modelle können dann Erkenntnisse für weitere Experimente abgeleitet werden. Einige Beispiele werden in der Folge besprochen. Das Machine Learning ersetzt dabei nicht den Forschenden, sondern liefert Daten und Modelle, für die man in realen Laborversuchen lange Zeit und viel Geld benötigen würde. Hören Sie in die neue Folge rein, wenn Sie wissen möchte wie sich die Welt der KI mit der Welt der Werkstoffe verbinden lässt und wie die Labore der Zukunft aussehen können.

On top of all that, he also has made a video game (!) that teaches material science, and that his students can even use to get credit on his courses (!!):

Get in touch

We will happily guide you through the emerging space of data-driven material discovery. We look forward to learn from your experience and problem space.


Explore some of the materials optimised with Matterhorn.


Biosynthetic building material


World-leading manufacturer in after-glow materials.