CV

Full CV as a PDF via the icon on the right.

Contact Information

Name Arthur R. C. McCray
Professional Title Postdoctoral Scholar, Stanford University
Email arthurmccray95@gmail.com
Website https://arthurmccray.github.io

Professional Summary

Research scientist working at the intersection of machine learning and electron microscopy. I build self-supervised and physics-informed methods for inverse problems in scientific imaging (deep generative priors, implicit neural representations, and differentiable forward models) and ship them as open-source tools used across the field. I am currently developing multimodal vision transformer foundation models for electron microscopy, pretrained on large simulated diffraction datasets together with crystal structure and composition, as part of a $1.2M Deep Material Priors program with Toyota Research Institute. Earlier work produced >10× faster and 40% higher-resolution ptychographic reconstructions and is being transferred into semiconductor metrology workflows.

Experience

  • 2024 - Present

    Stanford, CA

    Postdoctoral Scholar: Ophus Group
    Stanford University
    • Proposed and demonstrated a self-supervised deep-prior framework for solving inverse problems in (S)TEM, now a reusable approach underpinning projects in ptychography, tomography, and crystallography. Supervise extensions of the method by three graduate students and two postdocs, with collaborations at SLAC, UC Berkeley, and the Technion.
    • Initiated and led the first application of this framework to electron ptychography, achieving a >10× speedup and 40% resolution improvement over state-of-the-art ML and iterative baselines.
    • Led the technical contributions behind a $1.2M award from Toyota Research Institute for Deep Material Priors, a program building physics-aware foundation models for materials characterization. Now developing its multimodal vision transformer models (diffraction, structure, and composition encoders with a fusion backbone), pretrained with masked autoencoding on large simulated datasets using distributed (DDP/FSDP) training.
    • Developed whole-pattern-fitting methods for improved strain mapping, now being adopted for semiconductor metrology workflows at Samsung.
    • Created quantEM, an open-source library for quantitative electron microscopy, and lead its ptychography and machine learning development.
  • 2024 - 2024

    Berkeley, CA

    Postdoctoral Scholar: National Center for Electron Microscopy (NCEM)
    Lawrence Berkeley National Laboratory
    • Developed machine learning methods for 4D-STEM analysis in the group of Colin Ophus; continued the same research program upon the group’s move to Stanford.
  • 2018 - 2023

    IL

    Graduate Research Assistant: Petford-Long and Phatak Groups
    Argonne National Laboratory / Northwestern University
    • Built a simulation and training pipeline for Lorentz TEM neural networks, enabling large-scale statistical studies of spin textures.
    • Developed a self-supervised method for quantitative magnetic phase imaging from a single image, replacing the conventional through-focal series.
    • Authored PyLorentz, an open-source Python package for Lorentz TEM simulation and phase reconstruction now widely used in the magnetic-microscopy community.
    • Designed and ran cryogenic and in-situ TEM experiments on van der Waals magnets.
    • Collaborated across 10+ groups at national labs and universities, contributing analysis and imaging expertise to publications.

Education

  • 2018 - 2023

    Evanston, IL

    Ph.D.
    Northwestern University
    Applied Physics
    • Thesis: Observing and Understanding Complex Magnetic Spin Textures with Lorentz Microscopy
    • Advisor: Amanda Petford-Long
  • 2018

    Northfield, MN

    B.A., cum laude
    Carleton College
    Physics
    • Graduated with Distinction for the Senior Integrative Exercise

Open Source Projects

  • quantEM: core developer and package creator
    • Open-source Python library for quantitative electron microscopy.
    • Developed the ptychography and machine learning modules; contributed core infrastructure (data structures, visualization, utilities, configuration) and tutorials.
  • PyLorentz: sole developer and maintainer
    • Widely adopted Python codebase for Lorentz TEM simulation, phase reconstruction, and ML-based analysis.
    • Sole author and maintainer since 2020, covering architecture, documentation, releases, and user support.
  • emdatabase: co-developer
    • Python package and index for downloading electron microscopy datasets and model weights from anywhere they are publicly hosted, so notebooks and published results can be shared and reproduced with their data.
    • Co-developed with Carter Francis; completed the implementation and its integration into the electronmicroscopy ecosystem.

Honors and Awards

  • 2025
    Postdoctoral Scholar Award
    Microscopy and Microanalysis Conference
  • 2022
    Student Scholar Award
    Microscopy and Microanalysis Conference
  • 2021
    Editors' Suggestion
    Physical Review Applied
  • 2019
    Best Poster
    IEEE Magnetics Society & Nanotechnology Council, Chicago Section
  • 2018
    Distinction for Senior Integrative Exercise
    Carleton College

Publications

Invited and Contributed Talks

  • Invited: Deep Priors and Self-Supervised Learning for Electron Ptychography and Beyond. Microscopy and Microanalysis, Milwaukee, WI, 2026.
  • Improving Iterative Ptychography with Generative Priors. MRS Spring Meeting, Honolulu, HI, 2026.
  • Invited: Improving Iterative Ptychography with Machine Learning. Advanced Photon Source ptychography study group, 2025.
  • Improving Iterative Ptychographic Reconstructions with Generative Neural Networks. Microscopy and Microanalysis, Salt Lake City, UT, 2025.
  • Deep Learning Model for Obtaining Spatially Resolved Pair Distribution Functions from 4D-STEM Data. Microscopy and Microanalysis, Salt Lake City, UT, 2025.
  • Invited: Solving Inverse Problems in TEM with Machine Learning. SLAC Battery Center Seminar Series, 2025.
  • ML-Enabled Single Image Magnetic Phase Reconstruction for Lorentz Transmission Electron Microscopy. Microscopy and Microanalysis, Cleveland, OH, 2024.
  • 3D Magnetization Reconstruction for Lorentz Microscopy using Differential Programming. Microscopy and Microanalysis, Minneapolis, MN, 2023.
  • Behavior of Bubble Lattices in Cr₂Ge₂Te₆. 67th MMM Conference, Minneapolis, MN, 2022.
  • Invited: Evolution of Skyrmion Lattice Order in the van der Waals Ferromagnet Fe₃GeTe₂. Microscopy and Microanalysis, Portland, OR, 2022.
  • Simulation-Trained Machine Learning Models for Lorentz Microscopy. Microscopy and Microanalysis, Portland, OR, 2022.
  • Understanding Complex Magnetic Structures with Simulation-Assisted Lorentz Transmission Electron Microscopy. Northwestern University Applied Physics Seminar, 2021.
  • Invited: Convolutional Neural Networks for Image Processing. Biomedical Engineering Seminar Series, University of Illinois Urbana-Champaign, 2021.

Posters

  • Accelerating Ptychographic Phase Reconstructions with ML-Enabled Phase Unwrapping. Microscopy and Microanalysis, Cleveland, OH, 2024.
  • Control of Magnetic Skyrmions in an Exchange Biased van der Waals Ferromagnet. Microscopy and Microanalysis, Minneapolis, MN, 2023.
  • Cryogenic Lorentz Transmission Electron Microscopy of van der Waals Materials. Joint Meeting of IEEE Magnetics Society & Nanotechnology Council, Chicago Section, 2022.
  • Tailoring Magnetic Skyrmions in Multilayers by Anisotropy Engineering. Joint Meeting of IEEE Magnetics Society & Nanotechnology Council, Chicago Section, 2019. (Best Poster)
  • Orientational Disorder in Epitaxially Connected Quantum Dot Solids. Microscopy and Microanalysis, St. Louis, MO, 2017.

Teaching and Mentoring

  • Graduate student mentor, Stanford University (2024 – present): Day-to-day technical supervision of graduate students on ML and microscopy projects.
  • Guest lecturer, Stanford University (Fall 2024, Spring 2025): MatSci 327: TEM Analysis and Simulation.
  • Graduate teaching assistant, Northwestern University (2019, 2022): Physics and Materials Science & Engineering departments.
  • Undergraduate teaching assistant, Carleton College (2016 – 2018): Physics department.

Outreach

  • Scientist of the Month (2020 – present): Volunteer scientist at the Lakeside School and the Evergreen School, Seattle.
  • Stanford Science Pen-Pals (2024 – present)
  • Mentorship Opportunities for Research Engagement (2019 – 2023): Northwestern University board member and REACH coordinator.
  • Northwestern Prison Education Program (2020 – 2023)
  • Future City mentor (2018 – 2020): Saint Athanasius Middle School.
  • Invited outreach talk (2022): “High School to Physics PhD,” Open Gears Seminar, Metea Valley High School.

Professional Service

  • Journal referee: npj Computational Materials; APL Machine Learning; Advanced Functional Materials; Microscopy & Microanalysis; Ultramicroscopy; Nanoscale Horizons; Journal of Applied Physics; Scientific Reports; IEEE Transactions on Magnetics; Digital Discovery.

Skills

Machine learning: vision transformers, multimodal foundation models, self-supervised learning, inverse problems, implicit neural representations, deep generative priors and diffusion models, CNNs, differentiable programming, iterative reconstruction and optimization
Software: Python (NumPy, SciPy, scikit-learn), PyTorch, CUDA and JAX (working knowledge), package design and distribution, testing and CI, open-source maintenance
Systems: GPU training and profiling, distributed training, data pipelines for TB-scale imaging datasets, HPC (NERSC)
Simulation: multislice electron scattering, micromagnetics (MuMax, OOMMF)
Microscopy: Lorentz TEM, HR(S)TEM, 4D-STEM, electron diffraction, cryo-TEM (LN₂ and LHe), in-situ biasing and heating, EDS, MPMS/SQUID, MOKE, XRR, SEM, AFM
Nanofabrication: FIB patterning and lift-out, photolithography, sputtering, 2D-material exfoliation