Technical Lead, Flashblade//EXA

Description

The company is listed on the NYSE and ended fiscal 2026 with $3.7 billion in revenue, including its first quarter above $1 billion. Growth is accelerating into FY27.

Its market spans hyperscalers, AI labs, the AI hardware supply chain, data platform providers and the wider AI ecosystem. This role leads performance work on the company's exascale disaggregated storage system for AI factories, working with engineering, product management and field sales.

Responsibilities

• Own the performance story for the exascale storage system, from design through to how it is positioned against competitors in HPC and hyperscale AI.
• Turn storage protocol knowledge (NFSv3 direct-to-DN, pNFS, S3) into performance requirements for the disaggregated data path, targeting GPU-starved reads, metadata storms and bursty HPC loads.
• Align the roadmap, sizing guidance and reference architectures with what multi-tenant GPU cloud and sovereign cloud operators need when they buy.
• Write rigorous architecture analyses comparing metadata scaling, durability trade-offs and GPU-direct paths, so field engineers can handle customer evaluations and proofs of concept.
• Lead end-to-end performance validation across storage, RDMA networking and compute, and build repeatable benchmark suites (MLPerf Storage, IO500, elbencho) that test public claims.
• Set technical requirements and validation methods for QoS, workload isolation and IOPS limits, so shared deployments can meet SLA commitments.

Requirements

• A track record in performance work on file, block and object storage, such as parallel filesystems, NFS and pNFS at scale, and S3-scale object stores.
• Solid knowledge of GPU cluster design: RDMA fabrics, GPUDirect Storage, checkpoint and restore behaviour, and finding bottlenecks across compute, network and storage.
• Skill with MLPerf Storage, IO500 and load tools such as elbencho, fio and vdbench, plus the ability to design tests that mirror real workloads.
• Experience with architecture-level competitive analysis and a firm grasp of multi-tenant resource concepts (QoS, rate limiting, workload isolation) in distributed systems.
• Able to discuss topics from debugging low-level network queue stalls to explaining the system clearly to customer evaluation teams.

Conditions

Based in the Santa Clara, CA office. You are expected to work on site under company policy, except during PTO, work travel or other approved leave.

The annual base salary range is $239,000 - $359,000 USD. Ranges depend on role, level and location; for roles open in several locations, the range reflects the labor market across them.

Applications are accepted on an ongoing basis until the role is filled, with no fixed deadline.

Benefits

• May be eligible for incentive pay and/or equity.
• Flexible time off, wellness resources and company-sponsored team events.