The AtomGPTLab, led by Dr. Kamal Choudhary at Johns Hopkins University, invites applications for a Postdoctoral Fellow position in the fields of atomistic simulations, machine-learned force fields, and artificial intelligence (AI). The successful candidate will lead the development of a computational platform that unifies first-principles methods, classical molecular simulations, and cutting-edge AI techniques including graph neural networks (GNNs) and large language models (LLMs) to accelerate experimental design and discovery of novel materials. The research spans quantum mechanics, statistical physics, and deep learning, and aims to enable AI-guided predictions of synthesizable and functional materials such as superconductors, catalysts, semiconductors, and energy-relevant compounds. The position is embedded in an interdisciplinary and collaborative environment with active interactions across experimental groups and national laboratories. Basic Qualifications or Specialized Certifications + A PhD in Materials Science, Physics, Chemistry, Chemical Engineering, Computer Science, or a related field. + Demonstrated experience in one or more of the following: Density Functional Theory (DFT), machine-learned force fields (MLFF), graph neural networks (GNNs), or large language models (LLMs). Extensive Knowledge In: + First-principles simulations with packages such as VASP, Quantum ESPRESSO, GPAW. + Machine-learned interatomic potentials (e.g., ALIGNN-FF). + Structure-property prediction using GNNs (e.g., ALIGNN,). + LLM fine-tuning and prompt engineering (e.g., HuggingFace, OpenAI, AtomGPT). Working Knowledge Of: + Workflow tools (e.g.,JARVIS-Tools, ASE) and HPC environments. + Software development in Python, Git-based version control, and Conda packaging. + Data integration and surrogate modeling using experimental and computational datasets. + Interdisciplinary collaboration and mentoring of students or junior researchers. Specific Duties & Responsibilities + Conduct high-throughput DFT calculations and manage large-scale materials datasets. + Develop GNN architectures for predicting materials properties from atomic graphs. + Train and deploy machine-learned force fields for MD simulations and rapid screening. + Fine-tune or pre-train LLMs for generation and analysis of materials structures, synthesis protocols, and characterization outputs. + Build pipelines for combining experimental and simulated data for inverse design. + Provide real-time computational feedback to experimental collaborators for synthesis and characterization. + Lead manuscript writing, conference presentations, and contributions to open-source repositories. + Mentor undergraduate and graduate students, and participate in grant proposal development. Additional Opportunities + Collaborate as Co-PI on interdisciplinary proposals. + Engage with experimental groups, national labs, and industry partners. + Participate in the development of open cyberinfrastructure (e.g., AtomGPT.org). + Attend international conferences and contribute to global research communities. + Access to cutting-edge computing clusters and experimental characterization tools. Salary: $62,232 - $75,564 Applicants should submit a curriculum vitae and three recent publications. Review of applications will begin in mid-August 2025. Job Type: Full Time The listed salary range represents the minimum and maximum Johns Hopkins University offers for this position, based on a good faith estimate at the time of posting. Actual compensation will vary depending on factors such as location, skills, experience, market conditions, education, and internal equity. Not all candidates will qualify for the highest salary in the range. Johns Hopkins provides a comprehensive benefits package supporting health, career, and retirement. Learn more: https://hr.jhu.edu/benefits-worklife/. Equal Opportunity Employer All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, or status as a protected veteran. EEO is the Law https://www.eeoc.gov/sites/default/files/2023-06/22-088_EEOC_KnowYourRights6.12ScreenRdr.pdf