AI Storage Architect Lead — Hybrid Work

Applied Methods Ltd; Milpitas (CA); Hybrid; USD 180,000 - 280,000; Full time

Benefits offered by this job

Medical, dental, and vision coverage

Retirement savings benefits

Paid vacation, sick time, holidays

Flexible working hours & hybrid option

Professional development resources

Job summary

Graphcore in Milpitas, California, seeks a Principal Storage Architect to define high-performance storage architecture for AI training and inference workloads. You will lead the design of local and distributed storage tiers, optimize data paths among CPUs, accelerators, and NVMe devices, and drive the storage lifecycle and telemetry strategy.

You will collaborate with hardware, software, and systems teams to evaluate new storage technologies, ensure resilience, and achieve predictable tail

Qualifications

• Extensive storage engineering and architecture experience in AI/large data centers.
• Deep knowledge of NVMe, PCIe, NAND flash behavior, and device lifecycle management.
• Strong Linux storage internals and filesystem experience.
• Experience with storage benchmarking and workload modeling.
• Programming and automation in Python and Bash for monitoring and lifecycle workflows.

Responsibilities

• Define end-to-end storage architecture for local, disaggregated, and distributed storage.
• Profile and tune Linux storage stack to improve IOPS and latency.
• Lead storage control-plane and data-plane design across platforms.
• Set telemetry, alerts, and dashboards for fleet-scale storage health.
• Collaborate with cross-functional teams to select and deploy storage components.

Skills

NVMe

PCIe Gen5/Gen6

Linux storage internals

I/O schedulers

XFS/ext4/ZFS

fio

blktrace

iostat

perf

Python/Bash automation

kernel/driver debugging

Telemetry & alerting

Leadership & cross-functional

Education

Bachelor's or equivalent in CS/CE/EE

Tools

fio

blktrace

iostat

perf

Job description

Graphcore in Milpitas, California, seeks a Principal Storage Architect to define high-performance storage architecture for AI training and inference workloads. You will lead the design of local and distributed storage tiers, optimize data paths among CPUs, accelerators, and NVMe devices, and drive the storage lifecycle and telemetry strategy.

You will collaborate with hardware, software, and systems teams to evaluate new storage technologies, ensure resilience, and achieve predictable tail