Advanced Computing Engineer

Oak Ridge National Laboratory•Oak Ridge, TN

About The Position

The Advanced Computing Engineer designs, operates, and maintains high-performance computing (HPC) clusters and AI systems supporting research at Oak Ridge National Laboratory (ORNL). The position is part of the Emerging Technologies & AI and Computing team in the Research Computing group within the Information Technology Services Directorate. The Emerging Technology Computational Group provides HPC and AI systems engineering, integration, and support for ORNL’s research community. Its work includes designing, deploying, optimizing, monitoring, and providing tools across clustered infrastructures ranging from small clusters to systems with more than fifty thousand cores. The team partners with ORNL research organizations and other computing groups to identify appropriate solutions, and collaborates with customers to design, implement, and maintain computing environments throughout their lifecycle. The engineer promotes HPC and AI services to researchers, architects and implements scalable AI/ML solutions, and helps ensure the availability, performance, scalability, and security of production systems. The role also involves automation, monitoring, performance optimization, and collaboration with program technical points of contact to install and tune scientific toolsets. The successful candidate is expected to support ORNL’s mission and core values, and foster a respectful and equitable workplace. A bachelor’s degree in a relevant discipline and at least eight years of aligned professional experience are required; an equivalent combination of education and experience may be considered. The role requires substantial Linux systems experience, familiarity with configuration management and automation, scripting proficiency, and advanced Linux server troubleshooting and administration. Experience with HPC, AI/GPU systems, cluster technologies, monitoring, networking, and scientific or highly technical environments is preferred.

Requirements

Bachelor’s degree in computer science, computer engineering, information technology, information systems, science, engineering, business, or a related discipline, and at least 8 years of aligned professional experience; an equivalent combination of education and experience may be considered.
At least 4 years of experience managing UNIX/Linux systems.
At least 2 years of experience with configuration management and automation tools such as Git, Ansible, or Puppet.
Moderate proficiency in at least one scripting language, such as Bash or Python.
Experience with advanced troubleshooting and system administration of Linux servers.

Nice To Haves

Active DOE Q, DOD Top Secret, or DOD TS/SCI clearance is strongly preferred.
Interest in innovation and identifying technologies and opportunities, with the ability to communicate their potential benefits to teammates and research partners.
Collaborative, upbeat approach and ability to build trust and credibility with research teams.
Understanding of multiple operating systems and cluster technologies.
Experience with Rocky/CentOS/RHEL, Ubuntu, or VMware.
Understanding of HPC platforms, including supporting users with SLURM job submissions and troubleshooting.
Understanding of vLLM or TensorRT.
Experience deploying and operating AI workloads on GPU systems.
Knowledge of inference optimization, quantization, batching, context management, and model serving.
Experience building and running containerized applications in HPC and AI environments.
Experience with deployment mechanisms such as Diskless, Warewulf, and traditional deployment using Cobbler, PXEboot, or Bright.
Experience managing NVIDIA or AMD GPU cluster systems for AI/ML or image processing.
Knowledge of networking fundamentals, including TCP/IP, traffic analysis, common protocols, and network diagnostics.
Experience with InfiniBand networks and diagnostics.
Experience using performance and diagnostic tools to benchmark, analyze, and tune systems, networking, and storage.
Experience with Grafana, CheckMK, Nagios, Zabbix, SolarWinds, Ganglia, or other network and device monitoring systems.
Previous experience in a government, scientific, or other highly technical environment.
Good documentation skills, including the ability to prepare simple documentation web pages.

Responsibilities

Promote HPC and AI clustered computing services to researchers working with large datasets or developing code.
Design and define end-to-end AI/ML solution architectures for scalability, performance, and alignment with ORNL strategic goals and architectural guidelines.
Define and implement AI/ML solutions consistent with a master data management strategy that supports data integrity, security, and accessibility.
Ensure production systems’ availability, performance, scalability, and security.
Use automation and monitoring solutions to reduce routine maintenance and identify opportunities to improve system management and performance.
Collaborate with program technical points of contact to install scientific toolsets and help tune their performance.
Align work and interactions with ORNL’s values of Impact, Integrity, Teamwork, Safety, and Service, and promote an equitable, respectful workplace.