Staff AI Storage Platform Architect

Submer; España; Híbrido; EUR 130.000 - 170.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Hybrid work model

Competitive compensation

International team diversity

Descripción de la vacante

Radian Arc is seeking a Staff‑level Storage Architect to design and operate high‑throughput AI storage for GPU‑driven workloads across edge and core deployments.

You will shape storage architecture supporting distributed inference, fine‑tuning, and large‑scale model training, balancing throughput, latency, resilience, and cost. This role blends architecture ownership with hands‑on delivery in a hybrid work environment.

Formación

• Experience designing and operating distributed storage systems for high‑performance compute environments.
• Proven expertise in storage for large‑scale AI inference/training, including dataset distribution and checkpointing.
• Deep knowledge of the Linux storage stack and AI workload data access patterns.
• Familiarity with CSI/Kubernetes storage integrations and modern storage fabrics.
• Experience tuning storage for GPU‑accelerated workloads and multi‑tenant environments.
• Proven leadership in cross‑team architectural initiatives.

Responsabilidades

• Design scalable AI storage architectures for edge and core deployments.
• Define storage strategies for distributed inference, fine‑tuning and training workloads.
• Architect across StorPool, local NVMe, and disaggregated storage (VAST/Weka).
• Develop reference architectures, standards, and reusable patterns for future deployments.
• Collaborate with compute/network teams to ensure storage supports evolving AI workloads.

Conocimientos

Distributed storage design

AI storage

Linux I/O stack

Kubernetes CSI

Storage observability

Automation (Python/Bash)

Leadership

NVMe storage

Herramientas

StorPool

VAST Data

Weka

Kubernetes

MinioFS

Rook Ceph

Descripción del empleo

Radian Arc is seeking a Staff‑level Storage Architect to design and operate high‑throughput AI storage for GPU‑driven workloads across edge and core deployments.

You will shape storage architecture supporting distributed inference, fine‑tuning, and large‑scale model training, balancing throughput, latency, resilience, and cost. This role blends architecture ownership with hands‑on delivery in a hybrid work environment.