Senior HPC DevOps Engineer

New

EPAMHPC and MLOps

Opportunity to work remotely within PolandFull-TimeSenior

Salary not disclosed

Apply NowOpens the employer's application page

Job Details

Have 5+ years of experience in HPC or DevOps engineering roles.
Know MPI, OpenMP, and multi-node GPU communication protocols such as NCCL and GPUDirect.
Have experience managing AWS GPU instance families, including P-series, G-series, and Trainium/Inferentia.
Have hands-on experience with AWS Capacity Blocks for ML, On-Demand Capacity Reservations (ODCRs), and Service Quota management.
Have experience deploying containerized environments using Apptainer/Singularity, Docker, or Enroot.
Understand I/O performance bottlenecks with distributed file systems such as Lustre, GPFS, BeeGFS, or AWS FSx for Lustre.
Be able to profile applications using NVIDIA Nsight or similar tools to locate memory and compute bottlenecks.
Have experience deploying or scaling cloud HPC workloads using EFA, ParallelCluster, and parallel storage such as FSx for Lustre.
Have English proficiency at a B2+ level.

Guide scientists and data teams in using the platform UI and running self-service workloads.
Advise users on capacity blocks versus on-demand usage, optimizing cost, quotas, and resource availability.
Maintain automated pipelines for infrastructure provisioning and platform service deployments.
Resolve job scheduling failures, cluster bottlenecks, and resource quota queries.
Collaborate with developer experience teams to improve documentation.
Monitor GPU utilization with engineering teams using tools such as CloudWatch or Prometheus.
Manage AWS GPU instance families and allocate block compute for ML training and inference pipelines.
Plan capacity and manage reservations to ensure compute availability.
Deploy containerized environments tuned for HPC and GPU pass-through.
Deploy and scale cloud HPC workloads using parallel storage and networking solutions.

View Full Description & ApplyYou'll be redirected to the employer's site