Computing Engineer Workflow Modernization; HPC Engineer

What You Will Do

This position is open for external candidates only to apply.

The HPC Division is seeking an engineer to modernize how our Workload Management and Consulting teams do their own work, using AI to make our tools, documentation, and support processes more effective. This is a role between infrastructure and end users - you'll build capabilities that make both our team and our users more productive.

This position will be filled at either the HPC Engineer 2 or HPC Engineer 3 level, depending on the skills of the selected candidate.

• Identify which of the group's existing processes - documentation generation and upkeep, troubleshooting guides - are good candidates for AI-assisted automation, and build that automation. This includes surfacing relevant system and knowledge base content to assist consultants investigating complex user issues.
• Enable AI workloads on LANL's HPC systems through investigation and deployment of relevant technologies, such as:
• Slinky, a next-generation collection of projects bridging HPC and AI workloads.
• RunAI, a scheduler for GPU-centric AI/ML workloads, including capacity sharing and integration with existing HPC resources.
• Develop and deploy tools on Open Shift, for both direct user consumption and internal division use, including infrastructure that gives users visibility into their workloads.
• Provide technical support and documentation for the tools and workflows you build, and collaborate with vendors and open-source communities on relevant scheduling, orchestration, and AI tooling.

The tools you build will serve two audiences: HPC users trying to get work done, and HPC staff trying to support them.

At this level, responsibilities include:

• Building and iterating on AI-assisted tools for documentation and support workflows, under guidance from senior staff.
• Assisting with evaluation, deployment, and support of Slinky and RunAI in test and production environments.
• Developing and deploying tools on Open Shift.
• Providing day-to-day technical support and troubleshooting for users of these systems.
• Contributing to documentation and knowledge base content, including using the tools you help build.
• Collaborating with the Workload Management and Consulting teams to identify workflow pain points worth solving.

This level applies standard principles and techniques, builds familiarity with new AI tooling and scheduling/orchestration technologies, and exercises independent judgment on well-scoped problems.

In addition to the above, this level includes:

• Technical ownership of the group's AI-assisted workflow modernization efforts - deciding what to build, how to build it, and how to evaluate whether it's actually helping.
• Designing more complex tooling that combines AI-assisted automation with observability into user and staff workflows.
• Technical ownership of significant pieces of the Slinky and/or RunAI evaluation and deployment effort, drawing on hands-on experience with HPC workload management to guide design and integration decisions.
• Root-cause troubleshooting of complex issues spanning scheduling, orchestration, and the internal tools you build.
• Representing this work in technical discussions with vendors, other HPC teams, and the broader HPC community.
• Mentoring junior staff and students on AI tooling, scheduling systems, and container orchestration.

This level requires deep technical expertise in applying AI tooling to real operational problems, alongside solid grounding in scheduling and container orchestration.

LANL's High Performance Computing Division supports the Laboratory's mission by managing a world-class supercomputing center, supporting stockpile stewardship for…