# Principal Scientist, Predictive Sciences **Company**: GSK **Location**: King of Prussia, PA **Work arrangement**: hybrid **Experience**: senior **Job type**: full-time **Category**: Data Science **Industry**: Healthcare **Apply**: https://gsk.wd5.myworkdayjobs.com/en-US/GSKCareers/job/USA---Pennsylvania---King-of-Prussia/Principal-Scientist--Predictive-Sciences_449189?utm_source=yubhub.co&utm_medium=jobs_feed&utm_campaign=apply **Canonical**: https://yubhub.co/jobs/job_925dc95f-db7 ## Description We are building the next generation of biologics CMC with a rapid, efficient, and predict-first mindset. As a Principal Scientist in the Predictive Sciences group within our new CMC Digital and Knowledge Management department, you will sit at the intersection of late-phase CMC process development and design of cutting-edge digital tools, predictive solutions, and artificial intelligence that accelerate how we bring medicines from bench to patient. You will be an AI-native builder who takes mechanistic, hybrid and data-driven models from prototype to production. This role will deploy models through machine learning workflows and agentic AI systems that actively drive development decisions across upstream and downstream unit operations for biologics process development across our fed-batch, intensified fed-batch and continuous perfusion platforms. Strong modelling expertise is required and expected; equal weight is placed on the ability to ship those models into a form scientists can run independently, through deployment, product build and model lifecycle management. This role will also impact tech transfer, scale-up and commercialization of processes, and requires a proven ability to drive digital transformation principles across the business, with the agility to lead change. **Modeling & Simulation** - Build and deploy advanced process models: mechanistic, hybrid, and data-driven across upstream cell culture, chromatography, filtration, and formulation, including digital twins and PAT chemometric models for current and evolving continuous perfusion and capture platforms. - Drive model-informed process development and characterization as an embedded matrix team member, using MBDoE and model-based risk assessments to define critical process parameters, design space, in-silico experimentation, reduce wet-lab cycle, accelerate process understanding, platform fit, and model-based monitoring and process control strategies. - Translate science into software by converting mathematical models into working computational tools via platforms (gPROMS, SIMCA, JMP) or custom code (Python, MATLAB), and ensure models are validated to meet GSK internal compliance and regulatory requirements. - Champion adoption of digitally redesigned CMC workflows across the Drug Development and Supply organization, supporting the CMC portfolio through GSK enterprise-wide digital transformation principles, enhancing digital fluency, and embedding modelling as a central pillar of process development and tech transfers. - Build and sustain the modeling ecosystem – not just individual models, but the shared infrastructure, tools and ways of working that keep them versioned, monitored and reliable over time. - Stay at the forefront of the field through publications, conferences, and academic collaborations, bringing new techniques into the organization to continuously raise the bar on what's possible. - Communicate and collaborate broadly – present results to CMC teams and senior leadership, and foster strong working relationships across R&D, pilot plants, and the commercial supply chain. **Agentic AI based CMC Workflows** - Design and build agentic AI systems that autonomously parse historical CMC data, propose experimental designs and strategies, call specific models, and surface actionable recommendations to lab scientists and decision makers. - Integrate large language models and AI agents into CMC workflows – from literature mining and predictive intelligence to process insights, accelerated progress, and single cycle development. - Develop AI-driven risk assessment and criticality ranking tools to support process characterization and control strategy definition. **Data & Platform** - Work closely with laboratory informatics, data engineering, and tech platform teams to ensure data-linked digital solutions from electronic lab notebooks, process historians, and analytical instruments. - Champion FAIR data principles and structured data capture practices across CMC development teams. - Contribute to shared modelling infrastructure, reusable components, and validated computational workflows. **Scientific Partnership** - Embed as a trusted scientific partner within CMC teams – translating biological and process complexity into computable problems and translating model outputs back into practical development decisions. - Present modelling results and AI-derived insights to senior stakeholders, regulatory teams, and cross-functional partners. - Author or co-author technical reports, model qualification packages, and regulatory submission sections as needed. **Basic Qualifications:** - PhD (or Master's with 2 years of industry experience) in data science, computational science, AI/ML, chemical engineering, biochemical engineering, computational biology, or a closely related field. - Experience building predictive models for bioprocesses, such as mechanistic (e.g., kinetic cell culture models, chromatography mass transfer models), hybrid, or data-driven and chemometric models. - Experience in Python (experience with scientific computing stacks such as NumPy, SciPy, PyTorch, or similar) or equivalent, including and not limited to MATLAB, gPROMS, SIMCA etc. - Experience with CMC development principles – process development, analytical characterization, scale-up, and the regulatory framework governing biologics (ICH Q8/Q9/Q10/Q11). - Experience applying machine learning to small-to-medium scientific datasets, with clear understanding of overfitting, uncertainty quantification, and model interpretability. - Experience taking models into production use – deployment, versioning, monitoring, and MLOps practice (testing, packaging, CI/CD) – sufficient to hand a model to non-modelling end users with confidence. - Experience building or working with agentic AI frameworks in a scientific or industrial context. ## Skills ### Required - Artificial Intelligence (AI) - Computational Sciences - Computer Programming - Data Analysis - Data Management - Data Modeling - Data Science - Predictive Modeling - Statistical Models - Python - MATLAB - gPROMS - SIMCA - Machine Learning - Biological Engineering - Chemical Engineering ### Nice to have - PAT (Process Analytical Technology) - Design of Experiments - multivariate data analysis (MVDA) - regulatory filings (IND, BLA, NDA) - digital twin platforms - process simulation tools --- Source: [Apply at gsk.wd5.myworkdayjobs.com](https://gsk.wd5.myworkdayjobs.com/en-US/GSKCareers/job/USA---Pennsylvania---King-of-Prussia/Principal-Scientist--Predictive-Sciences_449189?utm_source=yubhub.co&utm_medium=jobs_feed&utm_campaign=apply)