PhD in Data Infrastructure and AI Workflows for Self-Driving Laboratories

Accelerating the discovery of clean energy materials requires autonomous experimentation environments known as Self-Driving Laboratories (SDLs). These labs depend on robust data infrastructures that support automation, reproducibility, and integration with machine learning tools. At DIFFER, in close collaboration with our research partners, we are developing such infrastructure to support a remote physical SDL dedicated to AI-driven experimentation in energy materials.

The goal of this PhD project is to design and implement a machine-learning-ready data infrastructure to manage and structure experimental data generated by this remote SDL. This includes ensuring that data from synthesis, characterization, and analytical instruments is consistently captured, harmonized, and annotated in formats suitable for AI-assisted experimentation. The candidate will focus on developing standardized data schemas and processing pipelines for experimental outputs, implementing metadata management and provenance tracking to ensure transparency, and building interfaces for accessing structured data through machine learning tools and workflows.

As a secondary objective, the candidate will explore how the structured datasets can be used in basic machine learning tasks, such as trend identification, clustering, or dimensionality reduction. This will help evaluate the quality and readiness of the data infrastructure for more complex AI applications and provide initial insights into its potential to support future experiment design.

This interdisciplinary project offers a unique opportunity to advance clean energy innovation at the intersection of data engineering, AI, and energy materials research.