Postdoctoral Scientist – Multimodal Representation Learning for Predictive Biology

Role: Postdoctoral Scientist in Data Analytics & Computational Sciences focused on multimodal representation learning for predictive biology within Johnson & Johnson's Innovative Medicine division. Primary objectives include developing AI/ML models to analyze heterogeneous biological data (imaging microscopy, phenomics, transcriptomics, proteomics) for drug discovery insights. Responsibilities encompass designing, implementing, and validating advanced AI/ML solutions, extracting biological insights from large-scale data, and collaborating with cross-functional teams to advance algorithms and support project goals. The candidate will publish research findings and contribute to peer-reviewed literature. Qualifications require a Ph.D. in Electrical Engineering, Biomedical Engineering, Computer Science, or related fields, completed within the last 3 years or soon, with demonstrated expertise in deep learning, representation learning, multimodal biological data modeling (e.g., Cell Painting, imaging, multi-omics), and proficiency in Python and AI frameworks like PyTorch or TensorFlow. Experience with foundation models, advanced architectures (Transformers, CNNs), and biological applications such as microscopy and multi-omics is preferred. The role involves close collaboration across research teams, with an emphasis on innovative AI solutions for understanding cellular heterogeneity relevant to drug discovery. Location options include Cambridge, MA (preferred), Spring House, PA, Beerse, Belgium, or Madrid, Spain, with no remote work. The position offers a competitive salary ($92K–$111K), performance bonuses, and benefits aligned with U.S. standards.