PhD Position in Agentic AI for Materials Discovery

This PhD position at the Dutch Institute for Fundamental Energy Research (DIFFER) focuses on integrating agentic AI with laboratory automation to accelerate materials discovery. The candidate will design Python-based workflows that translate AI-generated material compositions into executable synthesis and characterization plans, ensuring reproducibility and traceability. The role involves 2/3 hands-on experimental work, including high-throughput synthesis with nanoparticle printers and characterization via micro-XRF and SEM. The project collaborates with an industrial partner specializing in generative AI for materials, aiming to close the gap between AI proposals and experimental validation. The successful applicant will contribute to a FAIR-aligned dataset and publish findings in top venues while completing a PhD within four years.

• Design and implement agentic AI-driven experimental workflows that convert AI-generated material compositions into executable synthesis and characterization plans, ensuring quality control and traceability.
• Conduct high-throughput synthesis of material libraries using nanoparticle printers, systematically controlling composition and deposition conditions.
• Characterize samples via micro-XRF composition mapping and SEM, developing reproducible Python pipelines to extract composition, homogeneity, and morphology metrics.
• Develop and maintain a structured, FAIR-aligned dataset of experimental outcomes (successful, partial, and failed) to support AI model development and assess material realizability.
• Integrate AI agents with project databases, analysis tools, and reporting systems to ensure end-to-end traceability and reproducibility of results.
• Write well-tested, reproducible research software and contribute to publications, conferences, and the defense of a PhD thesis within four years.