Process Data Scientist

h3Process Data Scientist /h3pRequisition ID: 19417 /ppLocation: Zug, CH, 6300 /ppHolcim is the leading partner for sustainable construction, creating value across the built environment from infrastructure and industry to buildings. Headquartered in Zug, Switzerland, Holcim has more than 48,000 employees in 45 attractive markets – across Europe, Latin America, Asia, Middle East Africa. Holcim offers high-value end-to-end Building Materials and Building Solutions, from foundations and flooring to roofing and walling – powered by premium brands including ECOPlanet, ECOPact, and ECOCycle®. Find out more at /ph3WHO IS HOLCIM? /h3pAs a global leader in innovative and sustainable building solutions, Holcim is enabling greener cities, smarter infrastructure and improving living standards around the world. With sustainability at the core of our strategy, we are becoming a net-zero company, with our people and communities at the heart of our success. We are driving circular construction as a world leader in recycling to build more with less. It’s all thanks to our 70,000 talented people around the world who are passionate about building progress for people and the planet through four business segments: Cement, Ready-Mix Concrete, Aggregates and Solutions Products. /ph3Who Are Holcim People /h3pHolcim people are passionate about finding better ways to build. They embrace innovation and improvement with a pioneering spirit. They work as trusted partners, creating better solutions and experiences for their customers, communities, and colleagues. /ph3Role Overview /h3pAs we continue to look for more sustainable ways to build, we need world-class talent to join our team. People who are passionate about sustainability, driven by curiosity, and keen to grow, learn, develop, and thrive in our high-performance culture. /ppWe are looking for an innovative Process Data Scientist, motivated to solve the bigger operational and energy challenges in our industry. Inspired to develop and deploy cutting-edge AI applications that enhance a new era of kiln and thermal efficiency, our Process Data Scientist is passionate about leveraging digital tools to optimize process stability, reduce fuel consumption, drive Alternative Fuel substitution, and curb CO2 emissions across our Plants of Tomorrow. /ppYou will work on the algorithm design, Machine Learning development and deployment, and implementation of various manufacturing process optimisation initiatives including our flagship solution P-PREDICT. /ppYou will also make sure these solutions integrate seamlessly our legacy and new Plant of Tomorrow technologies to realize our vision for sustainable cement manufacturing. /ph3Main Responsibilities /h3bMachine Learning Physics-AI Hybrid Model Design, Tuning, and Validation: /bulliDesign, configure, and validate hybrid Machine Learning algorithms combining 1st-principles process physics and adaptive AI loops for calciner and kiln optimization. /liliDevelop predictive soft sensors for real-time process variables, like fuel calorific value forecasting of alternative fuels and clinker quality predicting. /liliBuild multi-variable target recommenders for control loops of our kilns and other equipment /liliUse deep learning, time-series forecasting, and probabilistic models to predict operational hazards (such as ring formation risk, cyclone blockages, and pressure spikes). /li /ulbModel Lifecycle Management /bulliResponsible for the end-to-end lifecycle of models (including feature engineering, design, cloud/edge deployment, continuous tuning, and automated retraining pipelines). /liliWork with our vendors on the deployment and customization of their solutions as well as on development of Holcim proprietary solutions. /li /ulbExplainable AI (XAI) Operator Interface Integration /bulliDesign operational logic, reason codes, and metrics to be displayed on operator dashboards to ensure high explainability and user trust. /liliCollaborate closely with process engineers and plant operators to translate operational knowledge into accurate model boundaries. /li /ulbField Adoption Knowledge Transfer /bulliDrive global uptake and utilization of P-PREDICT and other PoT initiatives across the world, through structured plant onboarding, commissioning support, and feedback loops. /liliProvide coaching, knowhow transfer, and technical documentation for plant process engineers. /li /ulh3Your Profile /h3h3Level Of Education/qualifications Normally Required /h3ulliMS or PhD in Computer Science, Chemical Engineering, Data Science, Electrical Engineering, Systems Control, or equivalent fields. /liliC3.ai Certifications (C3.ai Data Science, C3.ai V8 Data Science/Application Development) or equivalent enterprise AI framework certification (Advantage). /li /ulh3Experience /h3ulli3+ years’ experience in applied Machine Learning and Data Science projects focused on heavy industry process optimization (cement, chemical, energy, or mineral processing preferred). /lili2+ years’ experience productizing and scaling ML models in production environments (both Cloud and EDGE). /lili2+ years’ experience in industrial change management, coaching, and technical knowledge transfer to guarantee full plant adoption of digital control systems (Advantage). /li /ulh3Technical Competencies /h3ulliApplied Machine Learning expertise: Anomaly detection, time-series forecasting, regression analysis, probabilistic modeling, supervised classification, and unsupervised learning. /liliStrong mathematical domain background: Linear algebra, calculus, probability/statistics, heat and mass transfer, and basic process control dynamics (PID/MPC). /liliProgramming ML Stack: High proficiency in Python (PyTorch, TensorFlow, Scikit-learn, Pandas, NumPy) and R/SQL; experience with prototype and production data pipelines. /liliData Warehousing, Data Lakes Scalable Data Systems: Practical experience working with modern data platforms, including BigQuery, cloud data warehouses, data lakes, and lakehouse architectures on platforms such as GCP or AWS. Ability to efficiently query, transform, integrate, and prepare large-scale datasets for analytics and machine learning. Experience with distributed data processing and scalable ML platforms is a strong plus. /liliTime-Series Industrial Data Platforms: Experience working with time-series databases and industrial data platforms such as InfluxDB and Seeq. Familiarity with integrating sensor, process, operational, and contextual data from multiple sources is desirable. /liliData Engineering Pipelines: Experience designing and working with ETL/ELT pipelines, data ingestion, data transformation, feature engineering, data quality, and production data workflows. Understanding of batch and streaming data architectures is a plus. /liliData Visualization Explainable AI: Hands-on experience with Jupyter, Grafana, Looker studio,Seeq, Seaborn, and other visualization tools. Experience building explainable AI solutions, including model interpretation, feature importance, anomaly explanations, and reason-code displays for technical and business users. /liliCloud Production ML: Experience deploying and operating machine-learning solutions in cloud environments. Familiarity with MLOps, model deployment, monitoring, version control, CI/CD, and containerization is a plus. /li /ulh3Benefits /h3ulliHybrid work contract /liliFamily Support: Subsidized child care and paid maternity/paternity leave /liliProfessional Growth: Company-sponsored training and professional development. /liliDiscounts and Perks: Subsidized meals, exclusive discounts, and company gifts for special occasions. /liliLong-service awards /liliEmployee referral bonus /liliPension and Savings: Access to a comprehensive pension fund and an employee savings scheme. /li /ulpHolcim is committed to a diverse and inclusive workplace. We support equal opportunities for everyone, regardless of race, national origin, gender, sexuality, disability, or age. /ppImportant Notice: Holcim recruiters only contact candidates via official @holcim.com email addresses. We do not use Gmail or any other non-Holcim domains to conduct recruitment processes. Please verify the sender address before responding. /p #J-18808-Ljbffr