Jaehyung Lee

PhD Student, Materials Science & Engineering, Johns Hopkins University

prof_pic.jpg

I am a PhD student in Materials Science & Engineering at Johns Hopkins University, where I build machine learning and high-performance computing tools that accelerate the discovery and design of new materials as a member of the Choudhary Research Group.

My research sits at the intersection of first-principles simulation (DFT), machine-learning interatomic potentials, and agentic AI. I work on a range of problems: predicting electronic properties across millions of crystal structures, screening and designing battery cathode materials, learning catalytic adsorption energies, and building AtomGPT-powered agents that let researchers run materials-discovery workflows in natural language. I also care about making these methods fast and reproducible at scale, from single GPUs to multi-GPU systems.

Before Hopkins, I earned an M.S. in Chemical Engineering from Columbia University and a B.S. in Chemical Engineering from Penn State, with prior research in machine-learning potentials for battery materials and DFT studies of catalytic metal oxides.

Feel free to reach out to me at jlee859@jh.edu or grab my CV.

News

Jun 22, 2026 Presenting a poster on SlaKoNet DB: A Cross-Domain Tight-Binding Database of Electronic Structure at the Electronic Structure Workshop (ESW 2026), University of Wisconsin–Madison.
Jun 17, 2026 Presented a poster on SlaKoNet DB: A Cross-Domain Tight-Binding Database of Electronic Structure at Artificial Intelligence for Materials Science (AIMS) 2026 at NIST, Gaithersburg, MD.
Jun 16, 2026 Our paper AGAPI-Agents is published in The Journal of Physical Chemistry Letters (doi:10.1021/acs.jpclett.6c00837). :tada:
Jun 02, 2026 New preprint: RamanGPT — bidirectional mapping between crystal structures and Raman spectra with graph neural networks and generative transformers (arXiv:2606.03764).
Apr 01, 2026 Presented AGAPI at the MRS Spring Meeting 2026 in Honolulu, HI. :ocean:
Mar 27, 2026 New preprint: Lessons Learned from the 2025 Agentic AI for Science Hackathon is on ChemRxiv.
Feb 01, 2026 Presented a poster on BatteryMat: A Machine Learning Accelerated DFT Framework for Screening Battery Cathode Materials at the NIST Quantum Matters in Materials Science (QMMS) Workshop in Gaithersburg, MD.
Nov 06, 2025 Helped organize the 2025 Agentic AI for Science Hackathon at Johns Hopkins University, benchmarking agentic AI systems for scientific reasoning.

Selected Publications

  1. JPCL
    AGAPI-Agents: An Open-Access Agentic AI Platform for Accelerated Materials Design on AtomGPT.org
    Jaehyung Lee, Justin Ely, Kent Zhang, and 3 more authors
    The Journal of Physical Chemistry Letters, 2026
    DOI arXiv Bib
  2. Preprint
    RamanGPT: Bidirectional Mapping Between Crystal Structures and Raman Spectra with Graph Neural Networks and Generative Transformers
    Frank M. Abel, Jaehyung Lee, Charles R. Campbell, and 1 more author
    arXiv preprint arXiv:2606.03764, 2026
    arXiv Bib
  3. Preprint
    Lessons Learned from the 2025 Agentic AI for Science Hackathon
    Jaehyung Lee, H. Neralla, Charles R. Campbell, and 6 more authors
    ChemRxiv (under review, Mach. Learn.: Sci. Technol.), 2026
    Bib HTML
  4. Nano Futures
    The search for high-entropy fuel-cell catalysts using disorder descriptors
    Guangshuai Han, Tianhao Li, Xiao Xu, and 5 more authors
    Nano Futures, 2025
    DOI Bib