AI Storage Systems Architect for GPU-Cloud AI

Bitdeer; San Jose (CA); Hybrid; USD 140.000 - 210.000; Vollzeit

Zusammenfassung

Bitdeer Technologies Group seeks an experienced storage operations engineer to own and optimize AI training storage across our GPU-centric fabric. You will design parallel storage architectures, enforce multi-tenant QoS, and tune high-performance networks for low-latency data paths in a fast-moving AI environment.

You will instrument telemetry, drive predictors for storage faults, and convert incidents into automated remediation.

Qualifikationen

• 5+ years in enterprise or HPC storage operations, with at least 2 years supporting AI/ML workloads.
• Hands-on deployment and operations experience with WEKA, VAST Data, Ceph, or DDN/Lustre.
• Strong understanding of AI training I/O patterns: checkpoint frequency, dataset loading, and shuffle buffers.
• Experience with high-performance storage networking (NFS over RDMA, NVMe-oF).
• Knowledge of GPU Direct Storage and RDMA-based data transfer.
• Proficiency in storage performance benchmarking and tuning (fio, IOR, mdtest).
• Experience implementing multi-tenant storage with isolation and QoS.
• Strong Linux systems knowledge (kernel tuning, filesystem internals, block device management).
• Runbook-as-code mindset — executable SOPs within a quarter.

Aufgaben

• Deploy and operate parallel/distributed storage systems: WEKA, VAST Data, Ceph, DDN/Lustre; design AI-friendly storage architectures.
• Implement multi-tenant storage isolation with per-tenant QoS, quotas, and access controls; optimize GPU Direct Storage paths.
• Deploy and manage storage networking (NFS over RDMA, NVMe-oF) and Nvidia CMX for cluster-wide storage orchestration.
• Diagnose and tune storage performance: IOPS, throughput, latency profiling with fio, IOR, mdtest; maintain runbooks.
• Plan storage capacity aligned with GPU cluster growth and customer workloads; manage firmware, data migration, and DR procedures.
• Instrument storage telemetry — IO tail latency, checkpoint durations, NVMe SMART, filesystem health, RDMA counters.
• Partner with the platform team to define the storage-fault predictor and signal labeling.
• Convert incidents into automation: SOPs become runbook-as-code, runbook-as-code becomes agent remediation.

Kenntnisse

AI/ML storage patterns

Linux systems

Storage benchmarking

I/O performance tuning

Multi-tenant QoS

GPU Direct Storage

RDMA networking

Tools

WEKA

VAST Data

Ceph

DDN/Lustre

fio

mdtest

Jobbeschreibung

Bitdeer Technologies Group seeks an experienced storage operations engineer to own and optimize AI training storage across our GPU-centric fabric. You will design parallel storage architectures, enforce multi-tenant QoS, and tune high-performance networks for low-latency data paths in a fast-moving AI environment.

You will instrument telemetry, drive predictors for storage faults, and convert incidents into automated remediation.