How can autonomous systems remain reliable even when sensor data is inaccurate?This question has driven me for many years. During my PhD, I had the incredible opportunity to contribute a key piece to solving this puzzle: developing robust, adaptive algorithms that can adjust themselves in dynamic environments. Today, I am bringing this expertise to the industry, working at Siemens on an advanced radio-based positioning system that sets new standards for precise and robust location tracking in industrial automation.My strength lies in the application of probabilistic algorithms and sensor data fusion. Using techniques such as Bayesian inference and factor graphs, I develop systems that flexibly adapt to changing conditions – and remain reliable even when sensor data is corrupted by outliers. I have presented these approaches in numerous publications and conferences, and now I apply them to push the boundaries of autonomous systems in real-world applications.