Numerical Data Compression Postdoctoral Researcher

LLNL • Livermore, CA

1d

About The Position

Lawrence Livermore National Laboratory (LLNL) is seeking a Postdoctoral Research Staff Member to conduct fundamental research and development in numerical data compression. This role will support projects involving AI-based surrogate modeling, scientific computing, and the analysis of large datasets from physical and life sciences. The primary focus will be on advancing the state-of-the-art in lossy numerical data compression using tensor decomposition methods for multi-dimensional data. Key objectives include developing new coding schemes, number representations, and compact parameterizations for tensorial data; performing numerical analysis to characterize error distributions and establish error bounds; and creating highly scalable compression algorithms optimized for GPUs and multicore architectures. This position is within the Data Science & Analytics Group in the Center for Applied Scientific Computing (CASC) Division of the LLNL Computing Directorate.

Requirements

• Ph.D. in Computer Science, Mathematics, or a related field.
• Expertise in one or more of the following areas: data compression/reduction, information theory, (multi)linear algebra, or numerical analysis.
• Experience developing, implementing, and applying advanced algorithms to solve large-scale numerical or combinatorial problems.
• Experience with scientific programming in C/C++, CUDA/HIP/SYCL/OpenMP, Python, or similar, as evidenced through software artifacts.
• Demonstrated research productivity, as documented by publications, reports, presentations, and/or open-source software in relevant venues (DCC, TIT, TIP, SISC, SC, ISC, IPDPS, TVCG, VIS, etc.).
• Proficient verbal and written communication skills to collaborate effectively in a team environment and present and explain technical information.

Nice To Haves

• Experience with high-performance computing, GPU programming, parallel programming, and/or related methods including running numerical simulations involving complex workflows.
• Experience with (multi)linear algebra, including matrix and tensor decompositions.
• Experience working with large data sets and developing scalable solutions based on distributed-memory and/or out-of-core algorithms.
• Expertise in developing software prototypes using modern languages, libraries, and tools such as C/C++/CUDA/Python, Eigen/cuSOLVER/NumPy/PyTorch, git/CMake, etc.
• Familiarity with numerical compression methods.
• Familiarity with the basic principles behind machine learning.
• Skill set at the intersection of computer science and applied mathematics.
• Demonstrated technical leadership in fields related to computer science and applied mathematics, such as mentorship or team management.
• Experience with or interest in scientific applications such as fusion, earth system science, cosmology, seismology, materials science, medicine, etc.

Responsibilities

• Research, design, implement, and apply advanced numerical data compression and/or tensor decomposition methods (Tucker, TT, CP, etc.).
• Make independent contributions to project thrusts on novel coding methods, error analysis, and performance optimization.
• Document results in technical reports and peer-reviewed publications.
• Collaborate with domain scientists to evaluate the effectiveness of lossy compression methods and their impact on accuracy, storage, and performance within scientific workflows.
• Contribute to grant proposals and collaborate with academic and industrial partners.
• Pursue independent research interests and interact with internal and external scientists.
• Perform other duties as assigned.

Benefits

• Flexible Benefits Package
• 401(k)
• Relocation Assistance
• Education Reimbursement Program
• Flexible schedules (depending on project needs)

What This Job Offers

• Job Type: Full-time
• Career Level: Entry Level
• Education Level: Ph.D. or professional degree
• Number of Employees: 1,001-5,000 employees