Senior CORA HPC DevOps Engineer

We are looking for a Senior CORA HPC DevOps Engineer to lead the scaling, reliability, and automation of HPC CORA — the designated high-performance computing (HPC) and machine learning operations (MLOps) platform for Science, Innovation & Labs. EPAM Collaborative Omics Research Accelerator™ (CORA™) for AWS HealthOmics equips life science and healthcare organizations with an intuitive, secure, and GxP-ready environment to analyze and correlate omics data, utilizing AWS HealthOmics capabilities such as multiomic and multimodal analysis, population sequencing, and fully managed bioinformatics computation. Built for bioinformaticians, lab and bench scientists, and clinical providers, EPAM CORA offers a collaborative and extensible environment that democratizes access to high-performance omics computations. Responsibilities Guide scientists and data teams in navigating and leveraging the CORA user interface (UI) effectively, allowing them to execute self-service workloads without direct infrastructure friction Advise users and oversee infrastructure capacity across capacity blocks (reserved GPU/CPU capacity) versus on-demand usage, optimizing cost, quotas, and resource availability for heavy workloads Sustain automated CI/CD pipelines for infrastructure provisioning and platform service deployments Deliver Tier-2/3 support and troubleshooting for technical queries related to job scheduling failures, cluster bottlenecks, and resource quotas Work with developer experience teams to enhance platform documentation Team up with engineering groups to monitor GPU utilization (e.g., via CloudWatch/Prometheus) to reduce idle time, right-size clusters, and refine multi-node job scheduling Requirements 3+ years of experience in HPC DevOps or similar engineering roles Knowledge of MPI, OpenMP, and multi-node GPU communication protocols (NCCL, GPUDirect) Background in managing AWS GPU instance families (P-series, G-series, Trainium/Inferentia) and allocating block compute for large-scale ML training and inference pipelines Expertise in capacity planning and reservation using AWS Capacity Blocks for ML, On-Demand Capacity Reservations (ODCRs), and Service Quota management to secure compute availability Skills in deploying containerized environments tuned for HPC and GPU pass-through using Apptainer/Singularity, Docker, or Enroot Understanding of I/O performance bottlenecks when working with distributed file systems such as Lustre, GPFS, or BeeGFS Proficiency in profiling applications with NVIDIA Nsight or similar tools to identify memory and compute bottlenecks Competency in deploying or scaling HPC workloads on cloud infrastructure leveraging EFA, ParallelCluster, and parallel storage (FSx for Lustre) Intermediate English proficiency (B1+ level) Nice to have Familiarity with AWS Compute services Background in building and maintaining CI/CD workflows