YOUR AREA OF RESPONSIBILITY

In this role, you will primarily contribute to MaterialIQ, one of our key innovation projects. MaterialIQ focuses on developing AI systems based on material datasets to unlock deeper insights into material properties and behavior.

Your primary mission is to build internal AI and data science capability for material intelligence systems. You will work with structured, complex, and often limited material datasets and help develop robust, interpretable, and production-relevant models.

At the same time, you will ensure that knowledge about data pipelines, feature engineering, model assumptions, and validation strategies is understood, documented, and internalized.

WHAT SETS YOU APART

Responsibilities: 
• Work with structured material datasets and prepare them for AI and machine learning applications.
• Analyze and improve data pipelines, feature engineering approaches, and data representations.
• Support the development of predictive models for material properties and related use cases.
• Evaluate model performance with a strong focus on robustness, generalization, and uncertainty.
• Develop internal standards for model validation, interpretability, and reproducibility.
• Document system architectures, modeling decisions, assumptions, and limitations in a structured way.
• Collaborate closely with external partners and internal stakeholders to deeply understand AI systems.
• Translate AI and data science concepts into material-science-relevant insights.
• Contribute to building a scalable architecture for AI-enabled material data systems.
• Build and integrate AI-based tools and lightweight agent systems to improve workflows.
• Use LLMs pragmatically for automation, data processing, and developer support.
• Contribute to AI infrastructure and tooling design decisions.
• Ensure systems are reliable, maintainable, and usable in practice.

Your Profile

• Advanced English;
• Strong programming skills;
• Experience with machine learning in real-world settings, especially with structured or industrial datasets;
• Solid understanding of:
  • Feature engineering
  • Model validation
  • Handling small or complex datasets
• Ability to reason about model assumptions, limitations, and robustness;
• Experience with data pipelines and data preparation;
• Familiarity with modern AI tooling (LLMs, APIs, basic automation) — deep specialization in agent systems is not required;
• Exposure to scientific, engineering, or physics-related data is a strong plus.