Process Data Scientist: AI for Sustainable Cement (Hybrid)

Holcim Supplementary Pension Fund; Zug; Hybrid; CHF 140,000 - 190,000; Full time

Benefits offered by this job

Hybrid work contract

Family support

Professional growth

Discounts & perks

Long-service awards

Employee referral bonus

Pension & savings

Job summary

Holcim is seeking a Process Data Scientist to design hybrid ML models that blend first-principles process physics with AI. You will develop predictive sensors and support real-time decision-making to improve kiln and furnace efficiency, while ensuring seamless integration with existing plant technologies.

The role emphasizes end-to-end model lifecycle, explainability, and global knowledge transfer, including coaching plant engineers and driving adoption of P-Predict across PoT initiatives.

Qualifications

• MS or PhD in relevant technical fields as stated.
• 3+ years of applied ML and data science in heavy industry.
• Experience productizing ML models in production (cloud/edge).

Responsibilities

• Design, implement, and validate hybrid ML models (physics + AI).
• Develop soft sensors and predictors for real-time process variables.
• Lifecycle management including feature engineering and deployment.
• Collaborate with engineers to ensure explainability and adoption.

Skills

Machine Learning

Time-series forecasting

Python

Data Science

Anomaly detection

Explainable AI

Cloud & Edge deployment

Education

MS or PhD in Computer Science, Chemical Engineering, Data Science, Electrical Engineering, Systems & Control

Tools

PyTorch

TensorFlow

Scikit-learn

Pandas

NumPy

BigQuery

GCP/AWS

Job description

Holcim is seeking a Process Data Scientist to design hybrid ML models that blend first-principles process physics with AI. You will develop predictive sensors and support real-time decision-making to improve kiln and furnace efficiency, while ensuring seamless integration with existing plant technologies.

The role emphasizes end-to-end model lifecycle, explainability, and global knowledge transfer, including coaching plant engineers and driving adoption of P-Predict across PoT initiatives.