Post Doc - AI/ML and Optimization

Description

The global development data and AI group in the company's innovative medicine business is hiring a postdoctoral scientist focused on AI, machine learning and optimization.

The role applies machine learning, generative AI and optimization methods to computational tools for clinical operations, such as enrollment forecasting, cost estimation and optimization, and country and site selection.

Responsibilities

• Research, build, test and refine ML, GenAI and optimization pipelines for clinical development and trial operations.
• Prepare and analyze operational, real-world data and cost data to build ML models and optimization engines that predict outcomes, show trade-offs between competing trial goals, recommend the best operational scenarios and surface insights for early intervention and risk management.
• Build and assess LLM-based workflows for document extraction, protocol comparison, trial similarity, data harmonization, clinical concept interpretation and eligibility criteria review.
• Run stochastic simulations to forecast operational and patient journey outcomes, including enrollment and study completion.
• Design experiments and metrics that support validation, monitoring, data quality and responsible AI practices.
• Write reusable code, data pipelines and documentation with data scientists, engineers and clinical development partners.
• Explain research findings, methods and recommendations to technical and non-technical audiences, and contribute to scientific publication and presentations.

Requirements

• A Ph.D. in computer science, statistics, biostatistics, biomedical informatics, epidemiology, health economics, applied mathematics, engineering or another quantitative field, completed or finished before the start date.
• Hands-on experience with multi-modal ML predictive models and stochastic simulation for time-series forecasting.
• Experience with GenAI and clinical LLMs for document parsing, clinical concept disambiguation or harmonization.
• Experience building multi-objective optimization engines using evolutionary algorithms, reinforcement learning or mixed-integer linear programming.
• Proficiency in Python and SQL.
• Experience with Python LLM tools such as DSPy or LangChain, optimization tools such as pymoo, and ML tools such as Scikit-learn, XGBoost, Optuna or PyMC.
• Familiarity with clinical operational data, real-world data, electronic health records, claims and financial data.
• A track record of scientific communication through publications, conference talks, open-source work or comparable research output.
• Preferred: experience with EHR, claims, registry, CTMS, EDC or clinical trial data, plus privacy and data quality practices.
• Preferred: experience with MLOps tools such as MLflow and Kedro, Git, CI/CD stacks such as Jenkins and GitLab, and other DevOps tools.

Conditions

Hybrid work model.

Based in Cambridge, Massachusetts or Titusville, New Jersey, United States.

The company offers an inclusive interview process and treats each employee as an individual.