Description
The role partners with medical image reconstruction scientists and engineers to develop machine-learning components that improve reconstruction quality, speed, robustness, and quantitative accuracy. Responsibilities include defining training and evaluation pipelines, datasets, and metrics; productionizing models with inference performance, reproducibility, drift monitoring, and safe fallbacks; collaborating on hybrid physics-learned algorithms; and building experimentation and verification tooling. The position seeks strong applied machine-learning experience, signal-processing or imaging knowledge, a track record applying ML to physics-based or inverse problems, and experience with production systems, data curation, and possibly data assimilation.
