WEKA Data Platform Administrator (Remote US)

WEKA

5d•Remote

About The Position

The WEKA Data Platform Administrator operates WEKA deployments for AI hyperscaler customers as an extension of their teams, ensuring that tenants’ training, inference, and checkpointing workloads have a reliable data layer. The role manages deployments day to day and may work directly with customers’ tenants and platform teams to tune infrastructure for their workloads. The administrator manages deployment health, upgrades, capacity, configuration, audits, file systems, object-storage tiering, snapshots, quotas, and protocol settings. The role also optimizes infrastructure and client configurations, onboards tenants, establishes performance baselines, identifies scaling limits, and executes approved changes with validation, rollback plans, and documentation. The position maintains operational runbooks, turns repeated tasks into scripts or automation, prepares handovers, and explains procedures to customer operators. It also handles technical case escalation, communicates changes and risks to the wider WEKA team, collaborates with providers and customer-side teams, and contributes to weekly status updates and quarterly reviews.

Requirements

10+ years administering enterprise infrastructure at scale, with strong experience in file, object, or block storage.
Deep Linux administration skills, including RHEL-family and Ubuntu systems in distributed environments.
Hands-on experience with high-speed Ethernet and InfiniBand networking, RDMA, and fundamentals of DNS, ACLs, and TCP/IP.
Working knowledge of Kubernetes and containers, including persistent storage, CSI, and operators.
Experience operating GPU, HPC, or large cloud infrastructure for external customers or multiple tenants.
Experience with at least one major cloud: AWS, Azure, Google Cloud, or Oracle Cloud.
Proficiency in Python, Bash, or Ansible, and ability to automate repeated work.
Experience with monitoring and alerting stacks such as Prometheus, Grafana, and log and metrics tooling.
Experience with ticketing systems such as Jira and discipline in major incident management.
Clear written and verbal communication, including the ability to explain changes to engineers and executives.
Strong technical writing skills, including creating runbooks usable by someone unfamiliar with the environment.

Nice To Haves

Prior experience with WEKA or another parallel or distributed file system.
Experience running Slurm or Kubernetes-based AI/ML platforms.
Experience with NVIDIA GPU cluster networking and DGX-class systems.
Experience collaborating between support and product or engineering teams.
Familiarity with Confluence, Slack, and other collaboration tools.

Responsibilities

Operate WEKA deployments day to day, including monitoring, health, upgrades, capacity, configuration, and audits.
Create and manage file systems, object-storage tiering, snapshots, quotas, and POSIX, NFS, SMB, and S3 protocol configuration as tenants onboard and grow.
Maintain a healthy, documented, and recoverable operational state.
Execute approved changes by confirming scope and approval, classifying risk, validating, maintaining rollback options, and recording outcomes.
Monitor for configuration drift, capacity pressure, and anomalies, and raise concerns early with a recommended path.
Work with customer tenants and platform teams to understand storage use for training, inference, checkpointing, data loading, and scheduling.
Tune WEKA configurations for workloads, including filesystem layout, tiering, snapshot policies, quotas, and per-tenant QoS.
Validate and standardize client configurations, including GPU node clients, BlueField-3 and SR-IOV networking, mount options, cgroups and kubelet CPU policy, and the WEKA Kubernetes operator and CSI driver.
Onboard tenants and clients through repeatable, verified procedures; check configurations and scripts before production use.
Establish performance baselines, run benchmarks and tests with tenants, and report what changed and why.
Identify scaling limits, including client counts, process ceilings, and NAT and routing behavior across tenant networks.
Build and maintain customer operating runbooks covering procedures, access guidance, validation, rollback, and recovery.
Convert repeated manual work into scripts, automation, or checklists using Python, Bash, and Ansible.
Prepare handovers and backup briefs before time off or coverage changes.
Explain procedures and decisions to customer operators.
Open and update cases with observed conditions, customer impact, completed work, evidence, and requested next steps; follow escalation paths to Technical Services, GRE, and R&D.
Keep the extended WEKA team informed about changes, risks, and customer communications.
Work with cloud providers, network teams, and other third parties in customer environments to resolve issues.
Contribute operational results, trends, and risks to weekly status updates and quarterly reviews.