Hi, I’m Jason Schmidberger
I’m an Applied Scientist and Senior Data Scientist who enjoys tackling scientific and engineering problems that sit at the boundary between the physical world and data.
My career has taken me through environmental science, structural biology, computational biology, bioinformatics and, more recently, industrial sensing and climate technology. Although the application domains have changed, the underlying challenge has remained remarkably consistent: understanding complex systems from incomplete, noisy observations.
That challenge is what motivates me.
I enjoy developing analytical methods that combine scientific understanding with statistical inference to answer questions that cannot be addressed by data alone. Whether modelling enzyme evolution, reconstructing molecular structures, or estimating methane emissions from sparse sensor networks, my approach has always been to understand the underlying system first, then build models that respect both the physics and the uncertainty inherent in the observations.
Today I lead the development of Bayesian modelling algorithms for methane emissions monitoring at MIRICO, where I work on probabilistic source localisation, atmospheric transport modelling, uncertainty quantification and scalable analytical software. My interests increasingly lie in combining mechanistic models with modern statistical learning to build systems that remain interpretable, trustworthy and scientifically grounded.
More broadly, I’m interested in any problem where quantitative modelling can improve our understanding of complex natural or engineered systems. I don’t see myself as working exclusively in environmental sensing or life sciences; rather, I enjoy applying rigorous analytical thinking wherever difficult questions arise. The common thread throughout my career has been learning new domains quickly, extracting the essential structure of a problem, and developing computational methods that help others make better decisions.
My background in life sciences continues to influence how I approach data science. Biological systems taught me that complexity, uncertainty and imperfect observations are the norm rather than the exception. Those lessons remain just as relevant when modelling atmospheric transport, industrial processes or environmental measurements.
Areas of Interest
- Bayesian inference and uncertainty quantification
- Physics-informed and mechanistic modelling
- Environmental sensing and atmospheric transport
- Scientific machine learning
- Data assimilation and Bayesian state estimation
- Scientific software engineering
- Computational biology and bioinformatics
- Quantitative modelling of complex systems
Ultimately, I’m motivated less by any particular technology or application area than by the opportunity to understand complex systems and build models that make them more predictable, interpretable and useful.