AI Storage Architect for High-Performance GPU Cloud

Bitdeer (NASDAQ: BTDR)

San Jose (CA)

On-site

USD 180,000 - 240,000

Full time

2 days ago

Be an early applicant

Get more replies from employers

Send a job-specific resume in minutes.

Job summary

Bitdeer Technologies Group is seeking an experienced storage operations engineer to own and operate AI training storage across US data centers. You will deploy parallel storage systems, optimize checkpoints, and ensure scalable, fault-tolerant IO for 1,000-GPU workloads.

The role emphasizes automating SOPs into runbooks and feeding AIOps with actionable signals. You will collaborate with the platform team on storage fault predictors, telemetry, and remediation workflows, aiming to reduce MTTR

Qualifications

5+ years in enterprise or HPC storage operations, with AI/ML workload support
Hands-on deployment with two of WEKA, VAST Data, Ceph, DDN/Lustre
Strong understanding of AI training I/O patterns (checkpoints, dataset loading)
Experience with high-performance storage networking (NFS over RDMA, NVMe-oF)
Knowledge of GPU Direct Storage and RDMA data transfer
Proficiency in storage benchmarking/tuning (fio, IOR, mdtest)
Experience with multi-tenant storage isolation and QoS
Strong Linux systems knowledge (kernel tuning, filesystem internals, block devices)
Turn ops toil into ML signals; prior anomaly detector work is a plus
Runbook-as-code mindset; SOPs executable by machine

Responsibilities

Deploy and operate parallel/distributed storage systems (WEKA, VAST Data, Ceph, DDN/Lustre)
Design storage architectures for AI workloads and checkpoint I/O bursts
Configure per-tenant QoS, quotas, access controls; optimize GPU Direct Storage paths
Manage storage networking and cluster-wide storage orchestration
Tune storage performance: IOPS, throughput, latency with fio/IOR/mdtest; maintain runbooks