Principal Machine Learning Engineer (Reconstruction / Quantitative Imaging)

About the role

Partner with medical image reconstruction scientists and engineers to build ML components that improve reconstruction quality, speed, robustness, or quantitative accuracy. Define training/evaluation pipelines, datasets, and metrics that map to user needs and design requirements.

Productionize models: optimize inference performance, ensure reproducibility, monitor for drift/regressions, and implement safe fallbacks. Collaborate on hybrid algorithms incorporating physics and learned priors, denoisers, learned regularizers, and quality estimation.

Help build tooling for rapid experimentation as well as rigorous verification of algorithm changes.

Responsibilities

• Develop ML components to enhance medical image reconstruction (quality, speed, robustness, accuracy).
• Design and implement training/evaluation pipelines, datasets, and metrics aligned with user needs.
• Productionize models with focus on inference performance, reproducibility, and monitoring.
• Collaborate on hybrid algorithms combining physics-based and learned approaches.
• Build tooling for rapid experimentation and rigorous verification of algorithm changes.
• Leverage ML-based methods (e.g., PiNNs, Neural Operators) to solve PDEs in ultrasound simulation/imaging.
• Curate datasets from real-world sources, define ground truth, and manage labeling/simulation pipelines.

Requirements

• Strong applied ML experience with comfort in signal processing, imaging, or adjacent domains.
• Ability to transition between research prototypes and production-quality systems.
• Strong evaluation discipline: metrics, ablations, data leakage avoidance, and reproducibility.
• Track record of applying ML to physics-based or inverse problems (e.g., shipped projects, portfolio, publications).
• Experience with ML for imaging/inverse problems, including GPU performance constraints.
• Pragmatic production mindset: reproducible training/inference, regression testing, and safe deployment.
• Background in computational physics or scientific computing.

Skills

• ML for imaging/inverse problems with strong evaluation discipline.
• Data curation: building datasets from messy sources, defining ground truth, managing pipelines.
• Data assimilation: combining observations with physics-based models (Kalman filtering, variational methods, ensemble approaches).
• Agentic-SciML (a plus).