HPC Systems Engineer, Modeling & Simulation

The Lethality Modeling & Simulation team supports high-consequence engineering decisions through advanced modeling, simulation, and scientific computing. The team works across lethality, warhead effects, shock physics, detonation physics, solid mechanics, and related mission-critical analysis areas.

This work depends on reliable high-performance computing systems, specialized scientific codes, secure operating environments, and data workflows that allow analysts and engineers to move quickly. The team operates in an on-premises HPC environment and partners closely with engineering, analysis, IT, security, and specialized vendor partners.

Anduril is looking for an HPC Systems Administrator to support the computing environment used by the Lethality Modeling & Simulation team. This is not a traditional corporate IT administrator role. You will help administer and improve the on-prem HPC systems, scientific code environments, storage, runtime workflows, and tooling that enable analysts to run complex modeling and simulation workloads.

In this role, you will maintain and troubleshoot scientific computing systems, support Linux/Unix-based infrastructure, help manage storage and compute resources, and partner with analysts to resolve issues across code, nodes, queues, storage, and user workflows. You will also help develop tools, scripts, and data workflows that improve analyst productivity.

The ideal candidate has experience supporting HPC, scientific computing, research computing, or secure engineering computing environments. This role is scoped for a mid-level to senior individual contributor, with final level determined by depth of HPC experience, independence, technical ownership, and ability to lead improvements across the environment.

- Administer and support on-prem HPC systems used for lethality modeling and simulation workflows

- Maintain, install, update, and troubleshoot scientific codes and runtime environments

- Diagnose issues across code, compute nodes, operating systems, storage, queues, and user workflows

- Support Linux, Unix, and Windows administration across a specialized engineering computing environment

- Manage or support storage, data movement, and filesystem needs tied to modeling and simulation outputs

- Build scripts, automation, and tools that improve analyst productivity and reduce manual workflow friction

- Partner with analysts, engineers, security teams, and vendor partners to improve reliability, usability, and secure operation of the environment

- Support the team's path toward classified computing readiness

- Enable the Lethality Modeling & Simulation team to spend more time on mission-critical analysis and less time troubleshooting computing environments

- Improve the reliability and repeatability of scientific code execution across HPC systems, storage, and analyst workflows

- Help scale the team's modeling and simulation capability by building better tooling, automation, and data workflows

- Strengthen the team's secure computing foundation as Anduril expands into more sensitive and classified analysis environments

- 5+ years of experience administering Linux/Unix systems in HPC, scientific computing, research computing, defense, aerospace, national lab, or similarly complex technical environments

- Hands-on experience supporting compute clusters, scientific codes, engineering software, or modeling and simulation workflows

- Experience with Bash and/or Python scripting for automation, troubleshooting, or workflow support

- Experience with scientific codes or tools such as CTH, ALE3D, Sierra, ParaView, Cubit, VisIt, hydrocodes, solid mechanics codes, or adjacent simulation tools

- Experience with scientific software build/runtime environments, including tools such as CMake, compilers, MPI/OpenMPI, OpenMP, environment modules, or similar technologies

- Working knowledge of storage, filesystems, networking, and secure computing practices, including NFS/NAS, TCP/IP, trusted software sources, and data handling practices

- Bachelor's or master's degree in computer science, Computational Engineering, Electrical Engineering, Scientific Computing, or a related technical discipline

- Ability to work onsite and obtain and maintain a U.S. security clearance

- Must be a U.S. Person due to required access to U.S. export controlled information or facilities

- Current or prior U.S. security clearance; DOE, DoD, NNSA, or classified computing experience strongly preferred

- Experience in a National Laboratory, DoD research, defense, aerospace, or secure scientific computing environment

- Experience with Slurm or comparable HPC scheduling and cluster management tools

- Experience developing tools that improve analyst, researcher, or engineering productivity