Head of AI Infrastructure — Cluster SRE & Operations (Compute & Storage)

Nava; Bengaluru; On-site; INR 3,000,000 - 6,000,000; Full time

Job summary

Nava is building a leading neocloud platform for AI workloads in Bengaluru. The Compute & Storage SRE & Operations team ensures reliability of production NVIDIA GPU fleets and distributed storage, delivering resilient capacity and rapid recovery from faults.

You will own GPU compute and storage availability, drive hardware triage, automation, and incident response while coordinating with OEMs and platform engineering to improve fleet health and observability.

Qualifications

• 10–15 years in Linux production infrastructure, SRE or HPC with leadership experience.
• Hands-on GPU/server diagnostics, Linux internals and hardware lifecycle management.
• Operational knowledge of distributed/storage systems (WEKA, Lustre, BeeGFS, Ceph, IBM Storage Scale).
• Proficient with Python/Go automation, incident management and observability tooling.
• Familiarity with Kubernetes, Slurm and fleet-wide infrastructure operations.
• Experience operating large NVIDIA GPU clusters and NVLink/NVSwitch environments.

Responsibilities

• Own reliability and performance of production GPU compute and storage infrastructure.
• Lead hardware fault management, GPU health monitoring, incident recovery.
• Drive lifecycle management, automated remediation and service availability improvements.
• Improve observability with DCGM, telemetry, monitoring dashboards.
• Coordinate cross-functional incidents with OEMs, networking and platform teams.

Skills

Linux infra

SRE leadership

GPU clusters

Python automation

Kubernetes

Slurm

Tools

WEKA

Lustre

BeeGFS

IBM Storage Scale

Ceph

DCGM

Job description

About Nava

Nava is a neocloud company purpose-built for the AI era. We design, deploy, and operate large-scale GPU infrastructure and deliver inference-as-a-service to teams building the next generation of AI products. Our platform runs on NVIDIA GPU systems, high-performance RDMA fabrics, and a fully automated, software-defined operations model. Every engineer at Nava works close to the metal, on infrastructure built to keep GPUs saturated and models serving.

About Nava

Nava is a neocloud company purpose-built for the AI era. We design, deploy, and operate large-scale GPU infrastructure and deliver inference-as-a-service to teams building the next generation of AI products. Our platform runs on NVIDIA GPU systems, high-performance RDMA fabrics, and a fully automated, software-defined operations model. Every engineer at Nava works close to the metal, on infrastructure built to keep GPUs saturated and models serving.

About The Role

The Compute & Storage SRE & Operations team leads reliability engineering for production NVIDIA GPU fleets and distributed storage, ensuring customer workloads see healthy capacity, resilient storage and rapid recovery from infrastructure faults.

Own reliability and performance of production GPU compute and storage infrastructure. Lead hardware fault management, GPU health monitoring, incident recovery, lifecycle management, automated remediation, and infrastructure availability improvements.

What You'll Do

• Own GPU availability, node health, storage SLIs, capacity and service recovery objectives.
• Lead diagnosis of GPU Xid/ECC, PCIe, NVLink, firmware, OS and hardware faults.
• Own storage availability and performance investigations, recovery tests and lifecycle processes.
• Implement automated node isolation, hardware triage, repair/replace workflows and return-to-service checks.
• Improve observability using DCGM, Linux telemetry, service monitoring and fleet health dashboards.
• Lead cross-functional incidents and reliability improvements with OEMs, networking and platform engineering.

What You'll Need

• 10-15 years in Linux production infrastructure, SRE, HPC or distributed systems with people leadership.
• Hands-on GPU/server diagnostics, Linux internals and hardware lifecycle management.
• Operational knowledge of distributed or parallel storage such as WEKA, Lustre, BeeGFS, IBM Storage Scale or Ceph.
• Python/Go automation, incident management and observability systems.
• Familiarity with Kubernetes, Slurm and fleet-wide infrastructure operations.
• Within that, hands-on GPU / AI-datacenter exposure is what we weigh most - we value demonstrated depth over years, as the field is still emerging.
• Experience operating large NVIDIA GPU clusters and NVLink/NVSwitch environments.
• Hyperscaler or high-availability compute operations environments.

Skills: infrastructure,fleet,nvidia,reliability,storage,availability