Postdoc: Multimodal AI for Predictive Biology

Johnson & Johnson Innovative Medicine; Cambridge (MA); On-site; USD 92,000 - 111,000; Full time

Benefits offered by this job

Vacation – up to 120 hours per year

Sick time – up to 40 hours per year

Holiday pay – up to 13 days per year

Dental/Vision/Life insurance options

Job summary

Johnson & Johnson Innovative Medicine is recruiting a Postdoctoral Scientist – Multimodal Representation Learning for Predictive Biology to join the DDSAI organization for a two-year fixed term. The role focuses on AI/ML analytics for drug discovery, based at Cambridge, MA (preferred) with other locations available, no fully remote option.

The candidate will work at the intersection of advanced AI/ML modeling, multimodal representation learning, and predictive biology, contributing to

Qualifications

• PhD in Electrical Engineering, Biomedical Engineering, Computer Science, Applied Mathematics, or a related major (within 3 years of completion or to be completed).
• Strong technical expertise in AI/ML for biological applications with publications in top-tier AI/ML venues.
• Experience with foundation models and multimodal modeling of biological data (Cell Painting, imaging microscopy, multi-omics).
• Proficiency in Python and AI frameworks such as PyTorch or TensorFlow; knowledge of DL architectures (Transformers, CNNs, graph nets).
• Excellent communication and ability to present results in a multi-functional team.

Responsibilities

• Conceive, design, develop, implement and validate AI/ML solutions for drug discovery using multimodal data of perturbed cells.
• Analyze large-scale high-dimensional data to extract biological insights from imaging, phenomics, transcriptomics, and proteomics.
• Develop and evaluate computer vision solutions to quantify cellular heterogeneity.
• Collaborate with cross-functional teams to advance algorithms and product development.
• Communicate complex methodologies to diverse audiences and stakeholders.
• Maintain documentation and up-to-date code repositories.
• Draft manuscripts and disseminate findings in peer-reviewed venues.

Skills

AI/ML for biology

Multimodal modeling

Deep learning

Python

PyTorch

TensorFlow

Transformers

CNNs

Graph networks

Self-supervised learning

Multi-instance learning

Education

PhD in a related field

Job description

Johnson & Johnson Innovative Medicine is recruiting a Postdoctoral Scientist – Multimodal Representation Learning for Predictive Biology to join the DDSAI organization for a two-year fixed term. The role focuses on AI/ML analytics for drug discovery, based at Cambridge, MA (preferred) with other locations available, no fully remote option.

The candidate will work at the intersection of advanced AI/ML modeling, multimodal representation learning, and predictive biology, contributing to