AI Storage Engineer

We’re hiring an AI Storage Engineer to own the storage layer that keeps thousands of GPUs running at full throughput. This is a hands-on architect role: you’ll design the data path from ingest to checkpoint, run PoCs with storage OEMs, and make sure our NVIDIA Blackwell and Hopper platforms hit their designed performance.

What you’ll do

• Architect storage for large-scale AI training and inference clusters (DGX/HGX, SuperPOD, B300, GB300)
• Deploy and tune parallel file systems — WEKA, VAST Data, DDN EXAScaler, GPFS, Lustre
• Integrate GPUDirect Storage, NFS over RDMA and InfiniBand/RoCE fabrics
• Benchmark and optimise for LLM training checkpoints and RAG pipelines
• Lead vendor evaluations and PoCs with NVIDIA, VAST, WEKA, DDN and Pure Storage

What you bring

• 7+ years in enterprise storage, 2+ years in AI, HPC or GPU-cluster environments
• Deep hands-on expertise in WEKA, VAST, DDN, GPFS, Lustre or BeeGFS
• Working knowledge of InfiniBand/RoCEv2 and NVMe-oF
• Comfort operating at the intersection of storage, networking and GPU compute