Postdoctoral Research Associate (Machine Learning)

Location: On-site at the Upton, NY campus, with possible hybrid work arrangements. Moderate domestic and international travel is expected.

Salary: $71,900–$119,000 per year, commensurate with qualifications, education, experience, and internal peer-group considerations.

Company

hirify.global is a multidisciplinary U.S. Department of Energy laboratory conducting discovery science and transformative technology research across multiple scientific domains.

What you will do

• Design, develop, and deploy machine-learning and HPC workflows, algorithms, and software for DOE mission applications.
• Build software integrating machine-learning and numerical techniques for heterogeneous architectures, including GPUs, accelerators, and leadership-class supercomputers.
• Collaborate with computer scientists, computational scientists, applied mathematicians, and domain scientists to address scientific computing needs.
• Implement correctness and reproducibility testing, then analyze and optimize software scalability and performance.
• Apply software engineering and documentation practices to improve usability and maintainability.
• Present results at conferences and workshops and publish findings in conference proceedings or peer-reviewed journals.

Requirements

• PhD in Computational Physics, Chemistry, Materials Science, Computer Science or Engineering, Applied Mathematics, or a related field; the PhD must be obtained before employment begins.
• Strong experience developing, deploying, and optimizing applications and workflows in HPC environments.
• Programming proficiency in C/C++ and Python, with knowledge of at least one HPC parallel programming model such as MPI, OpenMP/OpenACC, CUDA, HIP, Kokkos, or SYCL/OpenCL.
• Hands-on machine-learning experience covering end-to-end model training, tuning, and evaluation.
• Understanding of DNN, CNN, transformer, and graph-based neural-network models and their scientific applications.
• Eligibility requires no more than five combined years of relevant postdoctoral and/or R&D experience after obtaining the PhD, subject to stated exclusions. The selected candidate must be able to obtain and maintain a DOE Uncleared Personal Identity Verification credential.

Nice to have

• Experience scaling machine-learning training or inference across multi-node HPC systems.
• Experience optimizing ML architectures and evaluating performance on CPUs and GPUs.
• Knowledge of scientific numerical algorithms, adaptive or agent-based workflows, and leadership-class or exascale platforms such as Perlmutter, Frontier, or Aurora.
• Contributions to collaborative or open-source software projects.

Culture & Benefits

• Collaborative, interdisciplinary research environment spanning computer science, mathematics, computational science, and scientific domains.
• Comprehensive employee benefits program.
• Initial two-year appointment with possible extension and career growth based on performance and funding.
• Work guided by integrity, responsibility, innovation, respect, teamwork, safety, and ethical accountability.