Postdoctoral Fellow, Multimodal Modeling

The Chan Zuckerberg Biohub Chicago, whose decoding-inflammation team builds tools for precise molecular measurements of inflammation in human tissues in collaboration with Northwestern University, the University of Chicago and the University of Illinois Urbana-Champaign, seeks a Postdoctoral Fellow in Multimodal Modeling for the launch of its Proteoform Spatial Biology Group, which studies the spatiotemporal regulation of proteins and proteoforms in inflammation and autoimmunity.

The fellow designs and trains self-supervised multimodal models that fuse confocal protein imaging, single-cell protein proximity networks and mass-spectrometry phosphoproteomics into shared representations (e.g. reconstruction and CLIP-style contrastive objectives), works with graph-structured proximity-network data, adapts imaging models for co-embedding, and uses the learned representations to model continuous cell states and the features driving state transitions.

The role is on site in Chicago with a base pay of $84,150 for a new hire.

PhD in machine learning, computational biology, bioengineering, biophysics or a related field; deep learning on images including pretrained vision models; supervised and unsupervised modelling; multimodal modelling or data fusion; graph neural networks or other graph representation learning; Python with PyTorch, JAX or TensorFlow. Nice to have: contrastive/self-supervised learning, topology-aware modelling, proteomics or mass spectrometry, single-cell omics, microscopy analysis, immunology.