Sr. Machine Learning Engineer, Medical Imaging

Intuitive•Sunnyvale, CA

About The Position

The Ion™ endoluminal system is Intuitive's new robotic platform for minimally invasive care in the peripheral lung, with an initial goal of improving the early diagnosis of lung cancer. We are seeking a senior algorithm engineer to play a lead technical role in conceptualizing, designing, and evaluating our next- generation planning and guidance software. This role is dedicated to developing advanced algorithms to analyze real-time ultrasound (US) imaging- specifically endobronchial ultrasound (EBUS) and min-profile ultrasound probes – for the accurate detection, segmentation, and classification pulmonary structures. This position focuses heavily on processing temporal, high-resolution ultrasound video streams, mitigating inherent acoustic artifacts, and fusing ultrasound data with pre-operative imaging modalities to improve targeting confidence and clinical outcomes. Drive the full cycle of medical imaging analysis software development, developing algorithms from R&D concepts into robust, commercial medical device products.

Requirements

MS or PhD in Computer Science, Electrical Engineering, Biomedical Engineering, or a related field.
5+ years of hands-on industrial experience in algorithm development, with a deep specialization in medical imaging and computer vision.
Exceptional proficiency in Python and C++, with a deep understanding of modern software component design for production environments.
Expert knowledge of ML frameworks and software toolkits, including PyTorch, OpenCV, MONAI, ITK, VTK, and model optimization tools such as TensorRT or ONNX.
Experience with spatiotemporal deep learning architectures and deployment of low-latency real-time applications for ultrasound imaging applications.
Excellent verbal and written communication skills for effective cross-functional collaboration with clinical and engineering teams.

Nice To Haves

Strong domain expertise in ultrasound physics, acoustic image processing, and temporal/video data analysis is highly preferred, especially experience with data from EBUS or high-resolution small-profile transducers.
Experience or knowledge of the development of AI/ML features in strict alignment with Good Machine Learning Practice (GMLP) principles, ensuring data representativeness and model transparency.
A proven track record of shipping commercial AI/ML software products is bonus.

Responsibilities

Design, prototype, and implement advanced computer vision and machine learning algorithms tailored for real-time processing of diverse ultrasound modalities, including EBUS and ultrasound probes.
Develop specialized algorithms to handle the unique challenges of high- frequency ultrasound data, including speckle reduction, acoustic shadowing mitigation, and temporal tracking across video frames.
Apply and fine-tune state-of-the-art architectures, including spatio- temporal Vision Transformers (ViTs), recurrent networks, nnU-Net, Graph Neural Network (GNN) and Diffusion models, adapting them for noisy, high-frame-rate clinical ultrasound datasets.
Architect and execute model deployment pipelines, seamlessly integrating trained models into high-performance C++ production environments, utilizing techniques like quantization, pruning, and hardware acceleration to ensure real-time performance on constrained medical systems.
Support system integration and testing while generating formal design documentation and patent applications.