Senior HPC Cloud Engineer: GPU-Accelerated Compute

Accenture-Federal-Services-2; Salt Lake City (UT); On-site; USD 110,000 - 224,000; Full time

Job summary

Accenture Federal Services in Utah is seeking a Senior Cloud Engineer / HPC specialist to design, size, tune, and operate high-performance compute clusters for mission workloads, including GPU-accelerated tasks.

You’ll own HPC scheduling (Slurm, PBS) and architect AWS ParallelCluster/AWS Batch for scalable provisioning, while collaborating with data platform teams to ensure secure, compliant operations. A Secret clearance is required, and related Agile practices are common.

Qualifications

• 5 years’ cloud engineering/infrastructure experience, including 2 years focused on HPC cluster design/operations.
• Experience with HPC job scheduling systems (Slurm, PBS, or equivalent).
• Proficiency with AWS ParallelCluster and/or AWS Batch for cloud-based HPC provisioning.
• Understanding of parallel computing paradigms (e.g.: MPI, shared-memory, distributed task orchestration).
• Working knowledge of DevOps practices (CI/CD, infrastructure-as-code, GitOps).
• Scripting proficiency in Python, Bash, or PowerShell

Responsibilities

• Design, size, tune, and operate HPC clusters for compute-intensive workloads.
• Own HPC job scheduling infrastructure (Slurm, PBS, or equivalent).
• Architect/manage AWS ParallelCluster and/or AWS Batch for cloud-based HPC provisioning.
• Design for parallel computing paradigms (MPI, shared-memory, distributed task orchestration).
• Configure industry GPU hardware/software for accelerated workloads.
• Tune CUDA, cuDNN, and GPU libraries for scientific computing/data processing.
• Integrate HPC compute with data pipelines (Kafka, Airflow, Spark) and AWS data services (EMR, Redshift, Glue).
• Ensure HPC infrastructure meets security controls/accreditation for classified environments.
• Document cluster architecture decisions, runbooks, and operational procedures

Skills

HPC cluster design

HPC operations

AWS ParallelCluster

AWS Batch

Slurm/PBS job schedulers

CUDA/cuDNN tuning

GPU-accelerated workloads

Kafka/Airflow/Spark data pipelines

Python Bash PowerShell

DevOps practices

Education

Bachelor's in Computer Science, Computational Science, or Engineering

Tools

AWS ParallelCluster

AWS Batch

Slurm

PBS

CUDA

cuDNN

Kafka

Airflow

Spark

EMR/Redshift/Glue

Job description

Accenture Federal Services in Utah is seeking a Senior Cloud Engineer / HPC specialist to design, size, tune, and operate high-performance compute clusters for mission workloads, including GPU-accelerated tasks.

You’ll own HPC scheduling (Slurm, PBS) and architect AWS ParallelCluster/AWS Batch for scalable provisioning, while collaborating with data platform teams to ensure secure, compliant operations. A Secret clearance is required, and related Agile practices are common.