Head of Platform Reliability

About the Role

Nava is building Asia’s next-generation AI Neo cloud platform. Backed by a recent $22M Series A led by Greenoaks Capital, we are rapidly scaling our AI-native GPU cloud. We are aggressively expanding our data center capacity, deploying multi-megawatt, high-density AI infrastructure across strategic Tier-1 APAC hubs to support massive-scale inference and training workloads.

We are looking for a Head of Reliability Engineering to own the end-to-end operational lifecycle and site reliability for our AI supercomputing clusters. This role uniquely bridges the physical and the virtual—requiring deep expertise in hardware operations, distributed systems, and modern SRE methodologies. You will ensure 99.99%+ availability across a complex stack spanning high-density facilities, bare-metal GPU nodes, InfiniBand/RoCE fabrics, and containerized orchestration layers.

What You Will Do

• Global SRE Leadership: Build, scale, and mentor a world-class team of Site Reliability Engineers, NOC analysts, and Hardware Operations specialists across the APAC region.
• Day 1 & Day 2 LCM Operations: Oversee the complete lifecycle management (LCM) of AI clusters—from Day 1 cluster bring-up, bare-metal provisioning, and extensive hardware burn-in/validation, to Day 2 continuous operations, zero-downtime firmware upgrades, fleet capacity scaling, and break-fix resolutions.
• E2E Infrastructure Health: Own the operational integrity of the entire AI stack—from data center facilities (power/cooling telemetry) up through container orchestration (Kubernetes) and AI workload schedulers (Slurm).
• Incident Command & Resolution: Establish rigorous incident response frameworks, 24/7 follow-the-sun on-call rotations, blameless post-mortems, and rapid mitigation strategies for complex hardware/software boundary failures (e.g., GPU memory errors, fabric congestion, driver kernel panics).
• Advanced Telemetry & Observability: Drive the strategy for high-cardinality observability, unifying hardware metrics (IPMI/Redfish, DCIM, thermal sensors, interconnect counters) with software telemetry (Prometheus, eBPF) for proactive fleet management.
• Reliability Engineering & Automation: Champion "Infrastructure as Code" (GitOps, Terraform, Ansible) to automate toil out of hardware lifecycle management, firmware flash pipelines, and cluster capacity scaling.
• SLO/SLA Management: Define, measure, and aggressively enforce strict Service Level Objectives (SLOs) and Error Budgets for multi-tenant customer workloads, ensuring maximum throughput and deterministic network performance.

What We Are Looking For

• Technical Breadth: Deep expertise spanning hyperscale data center operations, distributed Linux systems, high-performance networking (InfiniBand/RoCE), and modern SRE/DevOps practices. LCM & Operational Expertise: Proven experience executing seamless Day 1 deployments and sustaining robust Day 2 operations for massive-scale hardware fleets.
• Hardware/Software Symbiosis: Proven ability to manage the operational intricacies of high-density accelerator fabrics (NVIDIA GPUs, PCIe topologies, NVLink, advanced thermal management, and power distribution).
• Leadership Experience: A strong track record of leading and scaling global SRE, Cloud Operations, or Infrastructure Reliability teams in highly demanding, production-critical environments.
• Crisis Management: Calm under pressure with extensive experience in major incident management (MIM) and performing cross-stack root-cause analysis (RCA).
• Execution Focus: An absolute obsession with reducing Mean Time To Recovery (MTTR), maximizing Mean Time Between Failures (MTBF), and automating operational toil.

Ideal Background

Ideal candidates will have held titles such as Head of SRE, Director of Cloud Operations, Senior Manager of Infrastructure Reliability, or Head of Production Engineering at hyperscalers, GPU cloud providers, high-frequency trading firms, HPC organizations, or tier-one AI labs.

The Opportunity: This is a rare opportunity to define the reliability posture of a massively expanding AI cloud. You will have the autonomy to build an elite operations culture from the ground up, ensuring our infrastructure remains a rock-solid foundation for the world's most demanding AI workloads.