AI-Driven Cellular Physiology Modeling Postdoc

Berkeley Lab; Berkeley (CA); On-site; USD 99,000 - 111,000; Full time

Benefits offered by this job

Exceptional health benefits

Generous paid time off

Inclusive culture

Job summary

Berkeley Lab’s Biological Systems and Engineering Division seeks a Postdoctoral Fellow to join the Cellular Intelligence and BioAnalytics Group. You will develop AI-driven predictive metabolic models of cell physiology and metabolism and gain hands-on training in mammalian cell culture, micro-respirometry, stable isotope tracing, and biochemical assays.

The role includes collaboration with UCLA researchers, data analysis to advance predictive models, and publishing results in peer-reviewed

Qualifications

• Ph.D. completed within the last 1–2 years in a related discipline (CS, Chemical Engineering, Systems Biology, Bioengineering, Computational Biology).
• Strong computational and ML skills with experience applying AI/ML to complex biology datasets.
• Willingness to participate in hands-on wet-lab training onsite at UCLA to acquire cell culture, micro-respirometry, and biochemical assay skills.

Responsibilities

• Develop AI/ML models for predicting cell physiology, metabolism, and function.
• Collaborate with experimentalists to integrate UCLA and LBNL datasets into predictive models.
• Prepare and publish results in peer-reviewed journals and maintain thorough lab documentation.

Skills

AI/ML in bio

Cell biology knowledge

Python

Data analysis

Education

PhD in CS or Bioengineering

Tools

Python

TensorFlow

PyTorch

JAX

Job description

Berkeley Lab’s Biological Systems and Engineering Division seeks a Postdoctoral Fellow to join the Cellular Intelligence and BioAnalytics Group. You will develop AI-driven predictive metabolic models of cell physiology and metabolism and gain hands-on training in mammalian cell culture, micro-respirometry, stable isotope tracing, and biochemical assays.

The role includes collaboration with UCLA researchers, data analysis to advance predictive models, and publishing results in peer-reviewed