Postdoctoral Scientist - Rwe Neurology

Johnson & Johnson Johnson & Johnson · Pharma · Cambridge, MA +1

Postdoctoral Scientist role focused on developing and applying AI/ML methods to real-world evidence in neurology. The role involves analyzing patient data, evaluating model robustness and interpretability, and integrating multimodal data for clinical development and regulatory decision-making.

What you'd actually do

  1. Analyze real-world data sources, observational databases, registries, and historical clinical trial data to generate insights supporting neuroscience clinical development and regulatory evidence strategies.
  2. Develop and apply statistical and machine learning methods to support patient subtyping, disease-progression modeling, endpoint evaluation, treatment-pattern characterization, and clinical trial design.
  3. Evaluate model robustness and stability under clinically meaningful changes, such as cohort definitions, biomarker thresholds, endpoint assumptions, missingness patterns, or patient-selection criteria.
  4. Support development of interpretable and transparent modeling workflows, including model documentation, sensitivity analyses, bias assessment, and reproducibility checks.
  5. Integrate structured clinical data with selected multimodal inputs, such as biomarkers, clinical narratives, digital measures, or literature-derived features, where appropriate and feasible.

Skills

Required

  • Ph.D. or equivalent doctoral degree in biostatistics, statistics, epidemiology, data science, biomedical informatics, computer science, computational biology, bioengineering, or a related quantitative discipline.
  • Strong programming skills in R and/or Python, with experience analyzing patient-level healthcare or clinical data.
  • Experience with statistical modeling, machine learning, or causal inference methods applied to biomedical, clinical, or real-world data.
  • Hands-on experience with data extraction, cleaning, transformation, quality control, and reproducible analysis workflows.
  • Familiarity with methods to address confounding, selection bias, missing data, measurement error, or other sources of uncertainty in observational studies.
  • Strong written and verbal communication skills, with the ability to explain quantitative methods and results to multidisciplinary audiences.
  • Demonstrated ability to work collaboratively in a team-based scientific environment.

Nice to have

  • Experience with real-world data sources, such as electronic health records, insurance claims, patient registries, natural history studies, or other observational datasets.
  • Familiarity with clinical trial data structures, clinical endpoints, biomarker data, or regulatory evidence-generation workflows.
  • Experience with machine learning methods such as regularized regression, tree-based models, gradient boosting, survival models, representation learning, or explainable artificial intelligence methods.
  • Experience conducting sensitivity analyses, model validation, robustness checks, or reproducibility assessments.

What the JD emphasized

  • robust, interpretable, and fit-for-purpose
  • robustness
  • interpretable
  • transparent modeling workflows
  • reproducibility checks

Other signals

  • Develop and apply statistical and machine learning methods to support patient subtyping, disease-progression modeling, endpoint evaluation, treatment-pattern characterization, and clinical trial design.
  • Evaluate model robustness and stability under clinically meaningful changes
  • Support development of interpretable and transparent modeling workflows, including model documentation, sensitivity analyses, bias assessment, and reproducibility checks.