Talks and presentations

How to get from Sensor Data to an AI based Realtime Application - A technical Deep Dive

March 11, 2025

Talk, Embedded World Conference, Nürnberg

Which components does it need to create a predictive Maintenance application based on vibration data created by an industrial machine? That is a question that I get asked from many manufacturing companies that think about extending their traditional business by digital solutions. In this talk I try to answer that question leading through a showcase project and explaining each commponent of the project.

How MLOps Practices Speed Up the Development of Condition Monitoring Services at KSB SE & Co. KgaA

November 18, 2024

Talk, Machine Learning Week Europe, München

At KSB SE & Co. KGaA, the number of productive ML-models increased with the growing customer base of its digital product KSB Guard. As a result, the need for a clear separation of responsibilities between data scientists and operations increased. In particular, the further development of already productive models became a pain point. In this talk I presented a service architecture that promotes seamless collaboration between data science and operations teams.

Praxisbericht: ML Ops für Condition-Monitoring Anwendungen

April 24, 2024

Talk, Minds Mastering Machines, Köln

In this talk I worked out some talkeaways regarding MLOPS from one of our customer projects: Firstly, a clear understanding of the importance of MLOps for the long-term operation of machine learning models. Secondly, I gave Insight into which problems AWS Sagemaker can and cannot solve (model registry, multi-model endpoints, serverless endpoints,…). Finally, I conveyed an understanding of a service architecture that promotes seamless collaboration between data science and operations teams.

Data Science - Der Weg vom Sensor in die Cloud bis hin zur Endanwendung

December 07, 2022

Talk, 46. Deutsche Compoundiertagung 2022, Nürnberg

In this talk at VDI conference, I describe different views on Data and Data Science depending if you are a plant operators or a plant manufacturer. Furthermore, I give some insights about IT Infrastructure that brings sensor data from machines into the cloud in order to provide data driven services. Finally, I show my view on how Data Science is evolving in copounding technologies.

Kurz KI an der Pipeline

June 15, 2020

Podcast, KI in der Industrie, Virtual

In this podcast, I give some insights about a Data Science project I did at TAL Group. TAG Group is a pipeline operator that transports crude oil from the Mediteranean Sea all over the alps to different rafineries in Austria, Germany and Czech Republic. Therefore many pumps are operating all along the pipeline and in this project we identified inefficies and its root causes using Machine Learning.

Making Sense out of Sensor Data - Data Science at TAL-Group

May 12, 2020

Talk, Predictive Analytics World for Industry 4.0, Virtual Conference

In this talk, I give some insights about a Data Science project I did at TAL Group. TAG Group is a pipeline operator that transports crude oil from the Mediteranean Sea all over the alps to different rafineries in Austria, Germany and Czech Republic. Therefore many pumps are operating all along the pipeline and in this project we identified inefficies and its root causes using Machine Learning.

How Data Science was used to optimize energy consumption of transalpine oil pipeline

October 01, 2019

Talk, AIxIA – Artificial Intelligence meets Intelligence Artificielle, Karlsruhe, Germany

In this talk, I give some insights about a Data Science project I did at TAL Group. TAG Group is a pipeline operator that transports crude oil from the Mediteranean Sea all over the alps to different rafineries in Austria, Germany and Czech Republic. Therefore many pumps are operating all along the pipeline and in this project we identified inefficies and its root causes using Machine Learning.