AI/ML Engineer - Neuroclinical AI

Join us in pioneering breakthroughs in healthcare. For everyone. Everywhere. Sustainably.

Our inspiring and caring environment forms a global community that celebrates diversity and individuality. We encourage you to step beyond your comfort zone, offering resources and flexibility to foster your professional and personal growth, all while valuing your unique contributions.

Join Siemens Healthineers’ Digital Technology & Innovation (DTI) team in Princeton, NJ, a central hub for advancing Artificial Intelligence and Digital Technologies in healthcare.

As an AI/ML Engineer within our Neuroclinical AI team, you will develop AI algorithms and models that power image-guided neurovascular interventions, from procedure planning through real-time navigation in the angiography suite. Our primary clinical focus is acute ischemic stroke and intracranial aneurysm treatment, with other neurointerventional procedures also in scope.

You will work closely with clinical collaborators, imaging scientists, and product development teams to translate interventional workflow needs into robust, real-time AI solutions, spanning model development, validation, and deployment in a regulated medical device environment.

You are responsible for:

• Developing AI algorithms for real-time detection and tracking of interventional devices (e.g., guidewires, microcatheters, stent retrievers, coils, and flow diverters) in live fluoroscopy and angiography sequences
• Developing 2D/3D fusion and registration methods that bring pre-procedural and intra-procedural 3D imaging (e.g., CTA, MRA, 3D rotational angiography) and planning primitives such as vessel centerlines, aneurysm geometry, and target landmarks into the live 2D angiographic view
• Building foundation models for interventional and neurovascular imaging, and distilling them into compact, efficient models that meet real-time latency requirements on clinical hardware
• Building interventional digital twins and simulators of neurovascular anatomy, device behavior, and X-ray image formation for use in synthetic data generation and in training and evaluating device navigation agents
• Translating clinical requirements for neurointerventional workflows into appropriate data, annotation, and model performance requirements
• Performing validation and statistical evaluation of AI algorithms, including accuracy, robustness, and latency under realistic procedural conditions
• Coordinating activities across data collection, data annotation, model development, and validation throughout the project lifecycle, often across multiple concurrent projects
• Contributing to IP generation and peer-reviewed research in collaboration with clinical and technical stakeholders
• Proposing and driving new AI/ML-based initiatives in image-guided neurointervention
• Staying abreast of developments in the field and maintaining knowledge of state-of-the-art technologies and high-performance computing architectures

Minimum Qualifications:

• MS or PhD in Computer Science, Electrical Engineering, Biomedical Engineering, Robotics, or related field
• 2–5 years of experience in AI/ML development (industry or academic)
• Hands-on experience with Python, PyTorch or TensorFlow
• Experience developing and validating machine learning models for medical imaging or computer vision applications
• Strong problem-solving and analytical thinking skills
• Ability to clearly communicate technical concepts to both technical and clinical audiences

Preferred Qualifications:

• Experience with X-ray fluoroscopy, angiography, or cone-beam CT imaging, ideally in neurovascular or other interventional applications
• Experience with real-time object detection, segmentation, or tracking, particularly of thin or low-contrast structures in image sequences
• Experience with 2D/3D registration, multimodal image fusion, or 3D vascular modeling (e.g., vessel segmentation, centerline extraction)
• Experience with large-scale self-supervised pretraining or foundation models, and with efficient deployment techniques such as knowledge distillation, quantization, and optimized inference (e.g., TensorRT, ONNX)
• Experience with simulation, digital twins, or synthetic data generation (e.g., digitally reconstructed radiographs, physics-based device or blood-flow simulation)
• Experience with reinforcement learning, imitation learning, or robotic navigation, especially for autonomous or assisted catheter and guidewire navigation
• Familiarit…