Nicholas (Nico) Krämer

I am a machine learning researcher based in Copenhagen. I build algorithms for scientific machine learning, with a focus on fast and reliable uncertainty quantification. I've contributed to:

• Probabilistic numerical algorithms and probabilistic ODE/PDE solvers.
• Differentiable simulators and matrix-free linear algebra.
• Bayesian deep learning.

I write a bunch of code, mainly in Python/JAX (GitHub: @pnkraemer). For instance:

• Tueplots: Matplotlib styles for scientific papers. 750+ stars on GitHub, 250+ dependent projects.
• Matfree: differentiable linear algebra in JAX. 150+ stars on GitHub, 15+ dependent projects.
• Probdiffeq: probabilistic differential-equation solvers in JAX. 50+ stars on GitHub.

I am open to industry roles, starting Spring 2027 (Europe, ML research/engineering). CV available upon request. Contact info below.

Background

I am currently a postdoc in Søren Hauberg's group at DTU. Before that, I did probabilistic numerics in Philipp Hennig's group in Tübingen, and before that, scattered data approximation with Christian Rieger in Bonn. I studied Mathematics and Business Mathematics in Bonn and Mannheim. I am a member of the ELLIS Society and an affiliated researcher at the Pioneer Centre for AI (P1).

If you need a three-sentence biography, e.g. for announcing a talk, [click here].

Contact

I go by "Nico", but use "Nicholas" when writing papers. Online, I tend to be @pnkraemer.

• Google Scholar: click here
• GitHub: @pnkraemer
• LinkedIn: click here
• Email: pnkraemer (at) duck (dot) com or pekra (at) dtu (dot) dk

Talk to me in German (native), English, or Danish. I would love to hear from you!