Lead Applied ML Scientist

We’re looking for someone who can take an unsolved scientific ML problem, prove that the model works, understand where it doesn’t, and turn it into something an expert can trust.

The work sits at the intersection of deep learning, scientific imaging, physics, and uncertainty. Depending on your background, relevant experience might include:

• 3D / volumetric deep learning
• scientific or medical imaging
• physics-informed ML or inverse problems
• simulation-to-real learning
• self-supervised learning with limited labels
• generative models
• uncertainty estimation and model calibration
• PyTorch and research-to-production ML

You do not need to match every method above. What matters more is that you’ve solved difficult applied ML problems before, can identify and adapt the right approaches from the literature, and are comfortable owning a problem end to end.

This is a lead role for someone who can take scientific ML from research into a real product. You’ll set the technical direction, work closely with domain experts, and own the path from model development and validation through to a production system that users can actually trust.

Location: Remote. Switzerland-based candidates are especially welcome, but not required.