Staff Machine Learning Engineer

Job description

Job responsibilities

• Design, train, and evaluate deep learning models for semantic understanding of surgical scenes, including dense segmentation and structure detection from endoscopic imagery.
• Develop structured modeling and inference approaches using statistical modeling, optimization, and related algorithmic methods for complex real-world data.
• Develop machine learning components and supporting algorithms that meet real-time performance constraints.
• Own the end-to-end model lifecycle from research prototype to production: architecture design, large-scale training, model optimization (ONNX, TensorRT, mixed-precision), and integration with the da Vinci C++ software…
• Define evaluation methodology with clinically meaningful metrics and statistical validation frameworks appropriate for medical device regulatory submissions.
• Collaborate with surgeons, clinical scientists, and human factors engineers to translate clinical needs into technical requirements.
• Partner with systems and software engineering teams to ensure ML components meet real-time latency, memory, and reliability requirements for deployment on embedded robotic platforms.
• Mentor junior engineers and research scientists; establish best practices for experiment tracking, model validation, and reproducible research.

Job requirements

• M.S. or Ph.D. in Computer Science, Electrical Engineering, Robotics, Biomedical Engineering, Applied Mathematics, or a closely related field, with graduate training involving computer vision, machine learning,…
• 5+ years of post-Ph.D. or 7+ years of post-M.S. industry or applied research experience developing and shipping ML systems, with demonstrated impact on production-grade products or platforms.
• Deep expertise in deep learning for visual recognition, including semantic segmentation, object detection, and/or instance segmentation. Strong familiarity with modern vision architectures.
• Strong foundation in probabilistic modeling and Bayesian inference, including experience with one or more of: graphical models, nonlinear optimization, or MAP estimation for structured problems.
• Proficiency in 3D geometry and spatial reasoning: coordinate frame transformations, rotation representations (SO(3), quaternions, axis-angle), rigid and similarity registrations, and camera projection models.
• Hands-on experience with C++ development in a production context. Ability to read and navigate large C++ codebases, build system prototypes, debug C++ components, and interface ML models with C++ software stacks.
• Experience with model optimization and deployment for latency-sensitive applications: ONNX, TensorRT, quantization, mixed-precision inference, or equivalent embedded/edge deployment toolchains.
• Expert-level Python and PyTorch (or equivalent deep learning framework). Comfortable with NumPy, SciPy, and scientific computing at scale.