Machine Learning Engineer - Medical Imaging; Ultrasound

About

NodeAI is building AI to transform ultrasound and lymph node interpretation, grounded in real clinical workflows and rigorous validation.

We’re looking for a hands‑on ML Engineer to drive development of our ultrasound AI models across segmentation, classification, and object detection. You’ll design and train deep learning models, help set the technical direction for our imaging stack, and work closely with our engineering team to deploy them into a cloud‑based, clinically oriented product.

• Design, train, and evaluate deep learning models in PyTorch for ultrasound image analysis.
• Develop reusable models and shared architectures that support multiple tasks (segmentation, detection, classification), with a focus on efficiency and fast inference.
• Help set technical direction, identify and prioritize goals, and provide technical leadership/mentorship as the team grows.
• Implement scalable training and inference pipelines (multi‑GPU / multi‑node) and optimize/export models for deployment (e.g., ONNX, Tensor
  
  RT).
• Collaborate on MLOps workflows, including model versioning, monitoring, and safe rollout of model updates in a regulated medical environment.
• Help design labeling strategies (including active learning and QA of annotations) to efficiently improve model performance.
• Work with clinicians and product to define metrics and validation strategies for real‑world clinical use.

• Strong experience with PyTorch and modern computer vision architectures (e.g., UNet‑style models, transformers, detection frameworks).
• Experience with self‑supervised learning (SSL) and/or transfer learning to leverage limited labeled data.
• Solid software engineering skills:
  Python, Git, testing, CI/CD.
• Ability to collaborate with cloud/backend engineers (GCP or AWS, containers, Kubernetes) to integrate models into a production system.
• Comfortable working in a regulated environment and following best practices for data privacy, security, and reproducibility (HIPAA, audit trails, documentation).

• Experience with ultrasound or other medical imaging modalities and DICOM/PACS.
• Experience with real‑time or near real‑time image analysis and model optimization (pruning, quantization, distillation).
• Experience with Django and Django REST Framework for integrating ML models into production web applications.
• Familiarity with distributed task processing systems (Celery, Redis) for handling asynchronous ML inference and training workloads.
• Experience with real‑time video/image streaming, Web Socket communication, or processing live video feeds.
• Knowledge of edge computing, device synchronization, or multi‑server ML deployment architectures.

Based in Hamilton, Ontario, Canada, with in‑person work 3 days per week.

Competitive compensation package including equity incentives, with the opportunity to significantly influence both the product and the technical stack.