Senior Software Engineer (Deep Learning)

Germany, Munich

Our client is a venture-backed, fast growing deep tech startup operating in the medical imaging space.

Their mission is to reshape how clinicians interact with imaging data by developing intelligent, cloudbased AI solutions that support planning and decision making in critical interventions. Recently, their flagship AI platform received FDA 510(k) clearance and CE marking, and they are now scaling both our product and team. With an emphasis on scientific rigor, patient outcomes, and strong clinical partnerships, they offer the opportunity to work on challenging, high-impact problems in real-world healthcare.

The role

As a Senior Software Engineer in Deep Learning & Imaging, you will play a key role in designing, developing, and deploying production-grade AI models and quantitative imaging pipelines. You’ll collaborate with a multi-disciplinary team of software engineers, clinicians, product experts, and regulatory specialists to deliver clinically meaningful functionality at scale. This is a full-time and permanent role, with hybrid working.

Key responsibilities

Lead the design, implementation, and optimisation of deep learning models for image segmentation and analysis, particularly within the 3D medical imaging domain.

  • Build and refine quantitative imaging pipelines that derive clinically relevant metrics (e.g., organ volumes, anatomical geometry) from segmented data.
  • Manage the full model lifecycle, including data preparation, model training, validation, deployment, and monitoring.
  • Contribute to software architecture decisions, code reviews, and testing strategies in line with medical device standards.
  • Engage closely with clinical experts to define requirements and ensure medical validity of technical solutions.
  • Document software and algorithm designs, performance metrics, and processes to meet regulatory and quality standards.
  • Participate in continuous improvement of development practices, infrastructure, and team workflows.