Postdoctoral Fellow - Transmission Electron Microscopy, Texas Materials Institute, Cockrell School of Engineering

The Postdoctoral Fellow will lead research in AI-driven and self-driving transmission electron microscopy (TEM) as part of the advanced materials characterization and autonomous discovery initiatives within the Texas Materials Institute (TMI) at The University of Texas at Austin. This position focuses on developing the next generation of intelligent electron microscopy systems that integrate machine learning, robotic control, and real-time data analysis to achieve autonomous imaging and interpretation of complex materials systems. The Fellow will design and execute experiments that advance the frontier of self-optimizing microscopy, including automated alignment, adaptive focusing, drift correction, and AI-assisted atomic structure recognition. The role involves building and training deep-learning models for TEM image reconstruction and interpretation, linking image features to local chemistry, defects, and dynamic transformations under varying environmental or beam conditions. The successful candidate will work closely with faculty and research staff to help establish TMI’s new AI-integrated microscopy hub as a national leader in self-driving electron microscopy.

This position will operate as a core part of a larger AI-robotic materials discovery program at TMI that couples liquid-phase synthesis, high-throughput sample processing, and autonomous characterization. The TEM postdoctoral fellow will collaborate with companion postdocs specializing in liquid-phase synthesis and thin-film formation, integrating structural characterization directly with compositional and processing data streams. A central aspect of this work will be to interface TEM workflows with the microdroplet printing system for high-throughput sample deposition onto TEM grids or micro-electrode arrays, enabling statistically rich structure–property correlations across thousands of printed materials. The Fellow will also contribute to the development of robotic sample preparation and automated sample-loading systems, including a robot arm-based TEM grid handling and holder-loading setup, to achieve continuous, unsupervised operation of the microscope. Working within this integrated ecosystem, the Fellow will help connect real-time TEM data to cloud-based digital twins and the broader AI framework controlling synthesis and electrochemical testing, creating a closed experimental–computational feedback loop for autonomous materials discovery.

The Postdoctoral Fellow will be expected to lead and publish independent research, collaborate across disciplines of materials science, electron microscopy, chemistry, and data science, and contribute to the mentorship of graduate students and junior researchers. Additional responsibilities include developing new experimental protocols, contributing to multi-PI proposals, and helping define the architecture for fully autonomous electron microscopy systems. This position offers a unique opportunity to be at the forefront of AI-driven discovery science, operating at the interface between robotic automation, advanced electron microscopy, and intelligent materials design.

Responsibilities

• Develop and implement self-driving TEM workflows that integrate machine learning, computer vision, and automated microscope control for autonomous imaging, focusing, and data acquisition.
• Advance AI-assisted image interpretation, including atomic structure recognition, defect classification, and dynamic process tracking using deep-learning and physics-informed models.
• Integrate TEM operations with robotic sample handling, including the design, testing, and deployment of a robot-arm–based TEM grid-loading and exchange system for continuous, unattended operation.
• Collaborate with postdoctoral fellows in liquid-phase synthesis and microdroplet printing to establish seamless sample transfer pipelines from synthesis to TEM analysis, enabling high-throughput, correlative characterization.
• Develop and optimize sample preparation methods compatible with microdroplet-printed thin films, nanoparticle arrays, and electrochemical catalyst systems, ensuring reproducible and contamination-free data.
• Link real-time TEM data streams to digital twin and AI platforms, using cloud-based computation for adaptive experiment con