L3 Technical SME – AI Compute

L3 Technical SME – AI Compute

The Role

As an L3 Technical SME – AI Compute, you will be a senior technical authority for Spectrum Scale/GPFS and high-performance storage supporting AI and HPC environments.

You’ll bring deep technical expertise to some of the most complex infrastructure challenges, taking ownership of advanced troubleshooting, performance optimization, architecture decisions, and high-risk changes. You’ll work closely with engineering teams, customers, and technology partners to build secure, resilient, and high-performing storage platforms.

This is a hands-on role for someone who enjoys solving complex problems, sharing expertise, and continuously improving how enterprise-scale platforms are designed and operated.

What You’ll Do

• Serve as the senior technical escalation point for complex IBM Spectrum Scale/GPFS issues, including clusters, quorum, NSDs, failure groups, metadata, filesystems, and recovery.

• Lead deep root-cause analysis of I/O performance bottlenecks, intermittent path failures, and storage/network interactions, driving complex incidents through to permanent resolution.

• Design and validate Spectrum Scale architecture, including cluster topology, storage pools, filesets, policies, quotas, capacity, and performance baselines.

• Lead platform lifecycle activities, including patching, upgrades, compatibility validation, rollback planning, and high-risk change reviews.

• Optimize storage platforms for AI and HPC workloads, analyzing end-to-end throughput and latency across compute, network, filesystem, and backend storage layers, including GPUDirect Storage and RDMA where deployed.

• Define and continuously improve monitoring thresholds, operational standards, recovery procedures, disaster recovery testing, and L1/L2-to-L3 escalation criteria.

• Lead major incident root-cause analysis and corrective actions, engaging technology partners such as IBM and DDN when vendor-level expertise is required following comprehensive internal fault isolation.

• Provide technical leadership and mentorship to storage engineers, while reviewing architecture, implementation, commissioning, and acceptance evidence to ensure solutions meet performance and operational requirements.

Your Future at Kyndryl

Every position at Kyndryl offers a path to grow your skills and build your career. In this role, you’ll have the opportunity to deepen your expertise across AI/HPC infrastructure, high-performance storage, and enterprise-scale technology platforms, while working alongside experienced technical specialists and industry-leading technology partners.

As a senior technical SME, you’ll also have opportunities to influence architecture and operational standards, mentor other engineers, and contribute to increasingly complex and strategic customer environments.

At Kyndryl, we encourage continuous learning, collaboration, and technical growth, giving you the opportunity to expand your expertise and shape what comes next in your career.

Who You Are

You’re good at what you do and possess the required experience to prove it. However, equally as important – you have a growth mindset; keen to drive your own personal and professional development. You are customer-focused – someone who prioritizes customer success in their work. And finally, you’re open and borderless – naturally inclusive in how you work with others.

Required Skills & Experience

• 5+ years of hands-on IBM Spectrum Scale/GPFS engineering experience in large-scale, business-critical environments.

• Deep knowledge of GPFS architecture, including quorum, NSDs, failure groups, filesets, policies, recovery, and mm* tooling.

• Strong Linux, storage performance, and root-cause analysis skills, with proven experience resolving complex I/O, connectivity, and hardware issues.

• Solid understanding of SAN/NAS, Fibre Channel, iSCSI, NFS/SMB, and InfiniBand/RDMA, with the ability to independently resolve L3-level incidents.

Preferred Skills & Experience

• Experience supporting AI/HPC environments and optimizing high-performance storage for demanding workloads.

• Hands-on experience with DDN or comparable high-performance storage platforms.

• Experience with GPUDirect Storage, RDMA, and InfiniBand performance optimization and troubleshooting.

• Experience with capacity planning, DR/recovery testing, major incident RCA, vendor escalation, and mentoring L1/L2 engineers.