The Upstream Process Development group within the Bioprocess R&D organization is seeking a Postdoctoral Fellow – AI/ML Enabled Bioprocess Modeling and Control. The successful applicant will join a team of scientists and engineers focused on developing and optimizing manufacturing processes for recombinant proteins and other modalities for early- and late-phase human clinical trials. This role will focus on developing and applying innovative mathematical and computational modeling approaches to characterize, understand, and predict complex biological systems used for recombinant protein vaccine and therapeutic production. The postdoctoral fellow will develop hybrid mechanistic and data-driven models for mammalian cell culture processes and leverage transcriptomic and other omics data to enable early clone selection based on predicted process performance and generational stability. In addition, the individual will develop a model predictive control (MPC) framework that uses these models to enable real-time monitoring and control of cell culture processes. Further, the postdoctoral fellow will design and conduct targeted experiments to generate data for model development, training, validation, and control strategy evaluation. The role will also explore agentic AI approaches to orchestrate model fitting, transfer learning, and deployment across portfolio projects, enabling scalable and adaptive reuse of models for early decision-making and process control. This position is well suited for a highly motivated scientist with strong expertise in machine learning, systems biology, kinetic modeling, process control, and mammalian cell metabolism.

Develop hybrid mechanistic–data driven models for mammalian cell culture processes supporting recombinant protein production.

Integrate transcriptomic and other omics data as structured inputs for clone specific performance and stability prediction.

Apply machine learning and deep learning methods for phenotypic clustering, parameter estimation, and performance prediction.

Extend existing mechanistic bioprocess models to include additional physiological functions (e.g., amino acid metabolism, regulatory feedback loops) using kinetic, genome scale, or data driven modeling approaches.

Design and implement model predictive control (MPC) frameworks using mechanistic and hybrid models for real time control of critical process variables (e.g., feeding strategies, metabolite control).

Design and execute shake flask, ambr®, or bench scale bioreactor experiments to generate process and omics datasets for model development and validation.

Perform in silico sensitivity and scenario analyses to understand process robustness, control leverage, and drivers of performance and stability.

Validate models and control strategies using historical and new datasets, and deploy them prospectively to support new development programs.

Explore agentic AI frameworks to orchestrate model fitting, validation, and transfer learning across portfolio projects, enabling scalable adaptation of models to new clones and processes with human-in-the-loop decision support.

Maintain rigorous documentation in electronic laboratory notebooks and internal technical reports.

Communicate results effectively through presentations, technical discussions, and peer reviewed publications.

Collaborate with cross functional teams across different time zones and contribute to mentoring junior scientists as appropriate.