Research Computing Engineer (AI Infrastructure and HPC)

The Institute for Computational and Data Sciences (ICDS) at Penn State seeks a Research Computing Engineer to join our technical team. This role supports Penn State's research mission by designing, operating, automating, and optimizing the GPU and AI and computing infrastructure used by researchers across the university, along with the high-performance computing systems that support it.

The position will be filled at the Reserach Computing Systems Engineer - Advanced Professional level.

Candidates must be U.S. citizens due to specific access requirements associated with this position.

This position is ideal for an engineer who enjoys building reliable, scalable systems for machine learning, data-intensive research, and advanced computing workloads. The successful candidate will work across GPU systems, HPC platforms, storage, networking, automation, and user-facing research workflows to enable cutting-edge research in AI, simulation, and computational science.

Work Arrangement:This is a full-time position, which will report to the Research HPC Manager and requires on-site work at University Park and is not supportive of remote work.

Responsibilities: As part of a collaborative engineering team, you'll contribute across a broad range of responsibilities that include the following:

• Collaborate with teammates, users, and vendor support to diagnose issues and implement solutions across compute, storage, networking, and software environments.

• Monitor, maintain, automate, and improve AI and HPC systems and supporting infrastructure.

• Design, deploy, operate, troubleshoot, and optimize systems using DevOps and infrastructure-as-code practices.

• Support GPU-accelerated computing environments for AI, machine learning, and scientific workloads.

• Partner with researchers and ICDS staff to understand workload requirements and develop practical engineering solutions for system configuration, performance, and research workflows.

• Support security, logging, documentation, and compliance process for the systems we operate, including environments subject to federal research security.

• Contribute to planning, requirements gathering, process improvement, and operational readiness for new services and infrastructure.

• Provide timely updates to system documentation and respond to user questions with clear, actionable guidance.

• Evaluate and improve tools, platforms, and workflows that support AI model development, training, inference, and data movement at scale.

Required qualifications and skills include the following:

• Administration of multi-GPU nodes at scale, including driver and firmware lifecycle management, NVLink/NVSwitch topology validation, GPU health monitoring and tuning for multi-GPU or multi-node GPU workloads.

• Ability to work effectively in a Linux environment, including command-line tools, file editing, POSIX permissions, and system configuration.

• Strong scripting ability in Bash and Python.

• Strong problem-solving skills and the ability to debug complex systems.Ability to work collaboratively as part of a technical team.Experience using AI tools or AI agents to improve programming, debugging, development, or prototyping workflows.

• Clear written and verbal communication skills.

Preferred qualifications:

Experience with one or more of the following is helpful but not required:

• Experience supporting AI/ML infrastructure, including environments used for model training, inference, experiment workflows, and large-scale data processing.

• Experience with job schedulers such as Slurm, PBS, HTCondor, or LSF.

• Software development experience and familiarity with HPC programming environments such as C/C++, Fortran, CUDA, MPI, or OpenMP.

• DevOps experience, including Git-based workflows, CI/CD, automation, and collaborative development practices.

• Experience with system deployment tools such as xCAT, Warewulf, OpenCHAMI, OpenStack/Bifrost, or MAAS.

• Experience administering or supporting Kubernetes.

• Experience with virtualization and containerization technologies such as VMware, Docker, Apptainer, or Podman.

• Networking experience, including EVPN, BGP, and IPv6.

• Experience with high-speed interconnects such as InfiniBand, HPE Slingshot, or similar HPC fabrics.

• Experience with HPC or distributed storage systems such as GPFS, Lustre, Ceph, or VAST.

• Monitoring and observability experience with tools such as Grafana, Graphite, Prometheus, VictoriaMetrics, or InfluxDB.

• Experience with databases such as MySQL/MariaDB or PostgreSQL.

• Security experience including identity and access management, single sign-on, and LDAP/Active Directory.

• Familiarity with Agile project development.

• Prior experience in academic research computing, research data infrastructure, or large-scale shared computing environments.

Why ICDS

At ICDS, you'll build and support infrastructure that enables advanced research in artificial intelligence, machine learning, simulation, data science, and computationally intensive discovery. You'll work on clusters, storage and GPU systems Penn State researchers rely on for AI, machine learning and computational research. You won't own a narrow slice of that. Our engineers work across provisioning, scheduling, storage, networking, GPU infrastructure and user-facing tooling. You'll debug problems from a user's job submission all the way down to drivers and system firmware, and you'll learn a lot on the way.

You'll also be close to the science. You'll talk to the people running the workloads, understand what they're trying to accomplish, and engineer systems around actual research needs rather than requirements handed down from somewhere else. It's a collaborative team, and the problems are rarely the same twice.