Digital Innovation Engineer

Overview Celanese Engineered Materials is seeking an Engineer, Digital Innovation - Predictive Modeling & Advanced Experimentation role. This role is a specialized technical position focused on applying AI + Physics into predictive modeling, experimental design, and Bayesian optimization to enable faster, more confident decisions in new product and material development. This role is an opportunity to play a key role in advancing predictive modeling and advanced experimental strategy to accelerate the design and development of next‑generation materials. Applying rigorous quantitative methods to enable informed decision‑making early in technology and product development. The role operates at the intersection of modeling, statistics, and machine learning, with a strong emphasis on translating these capabilities into practical approaches that support technology and innovation programs. This position also builds and deploys digital methods to guide experimentation, prediction, and optimization that support computer aided engineering and new product development efforts. •Location can be hybrid in one of the following locations: Wilmington, DE Florence, KY Auburn Hills, MI Irving, TX Responsibilities Predictive Modeling for Material Property Design Develop and apply predictive and hybrid machine learning approaches for the prediction of properties key to designing the next generation of materials. Integrate mechanistic understanding, statistical modeling, and data‑driven methods to generate reliable, decision‑ready predictions. Quantify model confidence and limitations to support risk‑aware technical decisions. Translate complex modeling outputs into clear, actionable insights for technology and innovation stakeholders. Experimental Design & Bayesian Optimization for New Product Development Design and apply advanced experimental design strategies and Bayesian optimization for new product development. Efficiently explore high‑dimensional design spaces to prioritize experiments and identify optimal candidates for laboratory evaluation. Apply adaptive and sequential learning approaches to balance exploration and exploitation under limited data conditions. Qualifications Master's Degree or higher, or with equivalent experience in computer science, computer engineering, machine learning, physics, applied mathematics or related field Understanding of advanced materials, chemical processes, and laboratory data is a plus. 1+ years' work experience with modeling development, data analysis, business communication, and digital transformation is highly desirable. Proficiency in AI + physics-based machine learning. Working understanding of material science fundamentals Strong foundation in applied statistics, experimental design, and probabilistic modeling. Expertise in predictive modeling and simulation for material or system property prediction. Experience with uncertainty quantification, model validation, and decision support under uncertainty. Ability to translate advanced quantitative methods into practical workflows including proof-of-concept full-stack (backend + frontend) applications that inform technology and product decisions.

Working across the full lifecycle:

problem formulation → model and strategy development → application and adoption. Communicating complex modeling and experimental concepts clearly to diverse technical audiences. Influencing technology and innovation decisions through quantitative, model‑driven insight. Operating effectively in cross‑functional environments spanning product development, technology, innovation, and digital teams.