AI Storage Architect – High-Performance Data Centers

Graphcore; Milpitas (CA); Hybrid; USD 320,000 - 520,000; Full time

Benefits offered by this job

Hybrid working arrangements

Job summary

Graphcore in Milpitas, CA seeks a Principal Storage Architect to define high‑performance storage architecture for AI data center platforms and large‑scale training and inference workloads.

You will lead end‑to‑end topology design, tune Linux storage stacks, optimize NVMe/PCIe paths, and drive telemetry with cross‑functional leadership across hardware, software, and systems engineering.

Qualifications

• Extensive storage engineering and architecture for AI/HP data centers.
• Experience at Principal/Staff/Lead Architect scope.
• Knowledge of NVMe, PCIe Gen5/6 and SSD lifecycles.
• Strong Linux storage internals knowledge.
• Experience with benchmarking and profiling tools.
• Automation scripting in Python and Bash.
• Ability to lead cross-functional architecture decisions.
• Telemetry signals and alerting design experience.
• Root-cause analysis for storage issues.
• Vendor and partner collaboration experience.
• Evaluate new storage technologies and translate to platform requirements.

Responsibilities

• Define end-to-end storage architecture and roadmap.
• Design storage topologies across PCIe/network domains.
• Lead storage control-plane and data-plane architecture.
• Profile and tune Linux storage stack for latency.
• Optimize storage for AI workloads incl. checkpointing.
• Define telemetry and alerting for storage health.
• Lead root-cause analyses for storage issues.
• Collaborate with vendors and internal teams.
• Evaluate emerging storage technologies.

Skills

NVMe & PCIe

Linux storage internals

Storage benchmarking tools

Python & Bash

Telemetry & dashboards

Cross-functional leadership

Performance tuning

Troubleshooting complex systems

Storage architecture design

Education

Bachelor's or master's in CS/CE

Tools

fio

blktrace

iostat

perf

Job description

Graphcore in Milpitas, CA seeks a Principal Storage Architect to define high‑performance storage architecture for AI data center platforms and large‑scale training and inference workloads.

You will lead end‑to‑end topology design, tune Linux storage stacks, optimize NVMe/PCIe paths, and drive telemetry with cross‑functional leadership across hardware, software, and systems engineering.