Senior Engineer, Process Data Science and Statistics

AstraZeneca•Gaithersburg, MD

2d•$101,643 - $152,465

About The Position

The Senior Engineer will develop and maintain analytical workflows for multivariate process monitoring, predictive modeling of batch outcomes from incoming materials, and anomaly detection in high-dimensional process data. The role builds and maintains data systems, analytical platforms, and AI-enabled tools connecting process characterization and manufacturing execution data to statistical analysis and decision support. Work spans early process development through late-stage clinical development and commercial and post-commercial manufacturing. The role also supports hybrid mechanistic and data-driven models and human-supervised agentic AI applications. The position reports to the Director, Process Data Science and Statistics lead, Process Engineering, Cell Therapy Development and Operations. It provides statistical support for process and product characterization, comparability assessments, and control strategy development; develops process monitoring workflows, dashboards, predictive models, data pipelines, and AI/ML tools; and collaborates with R&D, MSAT, Manufacturing Operations, Quality, IT, and manufacturing sites. The role contributes to digital manufacturing capability development and shares statistical methods and reusable code across the organization.

Requirements

Ph.D. in Chemical Engineering, Biochemical Engineering, Biotechnology, Data Science, Statistics, or a related field and at least 2 years of relevant experience; or M.S. in one of these fields and at least 4 years of relevant experience; or B.S. in one of these fields and at least 5 years of relevant experience.
Demonstrated experience applying data science and statistical methods in process development or manufacturing, including process capability, variance components, DOE, control charting, and multivariate analysis.
Proficiency in Python, including pandas, scipy, statsmodels, scikit-learn, and plotly, and in SQL for data analysis and pipeline development.
Experience developing and evaluating predictive models, including data preparation, validation, and interpretation of model results.
Experience with software development practices, including version control, automation, and reproducible analysis workflows.
Ability to prepare clear technical documentation and translate complex statistical findings into actionable insights for cross-functional audiences.
Excellent cross-functional communication and collaboration skills in matrixed environments.
Ability to work independently, set priorities, and deliver under time-sensitive program and regulatory timelines.
Practical problem-solving skills and ability to translate analytical concepts into maintainable tools and workflows.

Nice To Haves

Experience working with cloud data systems such as AWS, Snowflake, Databricks, or similar.
Experience building interactive analytical applications such as Streamlit or Dash.
Experience with multivariate SPC methods, including PCA/PLS, Hotelling T-squared, SPE/DModX, and anomaly detection.
Experience with hybrid mechanistic and data-driven modeling, process simulation, digital shadows, or neural ODEs.
Experience with AI/ML methods applied to manufacturing or bioprocess problems, including incoming-material batch-outcome prediction and LLM-based or agentic tooling.
Experience in cell therapy or biologics manufacturing.
Experience with DOE design and analysis for process characterization.
Track record of building and deploying automated reporting and process monitoring systems.

Responsibilities

Support statistical analysis for comparability assessments, process characterization, and control strategy development for cell therapy programs.
Apply statistical frameworks such as equivalence testing, process capability analysis, variance component analysis, and tolerance intervals to evaluate drug product quality and support regulatory narratives.
Design and analyze DOE studies for process characterization and LVV manufacturing, including risk-based parameter selection and interaction modeling.
Support process and product characterization for early development activities.
Develop and maintain process monitoring workflows for pivotal and commercial manufacturing, including control charts, alert/action limits, and trend detection.
Build multivariate statistical process control capabilities to support continued process verification, batch review, and investigations.
Develop and evaluate anomaly detection methods for high-dimensional process data to support investigation triage and batch review.
Build interactive reports and dashboards for batch visibility, in-process trend tracking, and decision support across programs.
Develop and evaluate predictive models linking incoming material attributes to batch outcomes to support assessments before manufacturing run commitment.
Partner with IT to implement data pipelines and analytical platforms connecting process development and manufacturing execution data to statistical analysis and decision support.
Support manufacturing data digitization with IT and manufacturing sites to establish scalable data capture and integration for commercial readiness.
Develop and evaluate hybrid mechanistic and data-driven models, including digital shadows, for process understanding, forecasting, and decision support.
Build and maintain reproducible AI/ML workflows, including data preparation, model evaluation, documentation, and monitoring, with emphasis on explainability and traceability.
Develop constrained, human-supervised agentic AI applications for knowledge retrieval, investigation and deviation triage, digital batch-record assistance, scheduling support, and guided decision support.
Contribute to digital manufacturing and data science capability development, supporting AI/ML integration and scalable infrastructure for commercial readiness.
Translate process and manufacturing questions into practical analytical solutions, prioritizing multivariate process monitoring and incoming-material predictive modeling while supporting evaluation of emerging methods.
Share statistical methods, reusable code, and applied data science practices with scientists and engineers.
Collaborate with R&D, MSAT, Manufacturing Operations, Quality, and IT to align data capabilities with program needs.