Senior Machine Learning Engineer - Semantic Spatial AI (F/M/D)

The Opportunity

As a Senior Machine Learning Engineer, you will identify promising advances, rigorously evaluate them on our data, and make the strongest approaches work reliably at scale. Your primary focus will be point clouds and images, and in the future may expand to Gaussian splats, 360° video, CAD, and IFC data. You will configure, train, and adapt models, then integrate successful solutions into NavVis IVION to deliver measurable customer value.

This is a hands-on role at the intersection of computer vision, machine learning, and spatial data, working with one of the world's largest collections of high-fidelity 3D point clouds and registered 360° imagery of real built environments.

How You Will Make an Impact

• Evaluate and integrate AI/ML models, systems, and frameworks for spatial data, primarily point cloud and image data, applying them to use cases such as object segmentation, semantic scene understanding, and generating derived artifacts such as a floor layout plan.
• Own the practical details that determine whether a system works in production: understand its constraints, make sound architectural choices, configure it correctly, prepare and preprocess data, and optimize its performance on our datasets.
• Design systematic evaluations using rigorous methodology so decisions are driven by evidence rather than impressions.
• Diagnose, debug, and enhance ML systems when results fall short, applying a deeply analytical mindset to understand why a model behaves the way it does and how to improve it.
• Adapt or fine-tune existing models to our data where it delivers meaningful value, focusing on applying and getting the most from the state of the art rather than from-scratch model development and training.
• Work closely with your team of computer vision, computer graphics, data processing, and full-stack engineers to turn ML capabilities into valuable features our users can rely on.

What Will Help You Succeed in the Role

• A Master's or PhD in a relevant field (equivalent practical experience is equally valued).
• Several years of experience applying machine learning to real-world problems (exceptional candidates with less experience will also be considered).
• Proven ability to adapt ML models and frameworks to real data, including data preparation, configuration, and performance optimization.
• A rigorous, evidence-driven approach to evaluation and experimentation, paired with strong analytical and debugging skills to understand and improve complex ML systems.
• Strong Python skills and experience with modern deep-learning frameworks such as PyTorch or TensorFlow (C++ experience is a plus).
• A solid foundation in machine learning, linear algebra, and geometry.
• Experience with spatial data, especially point clouds, images, or video data (familiarity with relevant architectures such as point or vision transformers, Segment Anything models, or Grounding DINO/CLIP is a plus).