WEKA Data Platform Administrator (Remote US)

Role overview

Own the day-to-day operation of a distributed data platform used by GPU cloud providers and their tenants. You will keep storage services healthy, documented, and recoverable, while helping customers tune infrastructure for AI, machine learning, and other demanding workloads. The work combines hands-on administration, customer support, performance troubleshooting, and operational improvement.

Responsibilities

- Monitor platform health, capacity, configuration, upgrades, and audit readiness; identify drift, anomalies, and emerging capacity risks. - Configure file systems, object-storage tiering, snapshots, quotas, and access protocols as tenants are onboarded and grow. - Tune storage and client settings for training, inference, checkpointing, and data loading; establish baselines and explain performance changes. - Verify GPU client and Kubernetes storage configurations, networking, mount options, and resource policies before production use. - Maintain clear operating procedures with validation, rollback, and recovery steps; automate repeated work with scripts, checklists, or tools. - Coordinate incident cases and escalations with customer teams, cloud and network providers, and internal engineering groups.

Requirements

- At least ten years administering enterprise infrastructure at scale, including substantial file, object, or block storage experience. - Strong Linux administration skills across RHEL-family distributions and Ubuntu. - Practical knowledge of high-speed Ethernet or InfiniBand networking, RDMA, DNS, ACLs, and TCP/IP. - Working knowledge of Kubernetes and containers, including persistent storage, CSI, and operators. - Experience supporting GPU, HPC, or large cloud environments for external customers or multiple tenants, and at least one major cloud provider. - Scripting or automation skills in Python, Bash, or Ansible; familiarity with monitoring, incident management, and ticketing tools. - Clear customer communication and technical writing skills, including runbooks that another operator can follow.

Nice to have

Experience with parallel or distributed file systems, Slurm or Kubernetes-based AI platforms, GPU cluster networking, or close collaboration between support and engineering teams.