Senior Infrastructure Engineer — GPU & High-Performance Computing (Up to S$380K+, Urgent! SG-based)

**Senior Infrastructure Engineer — GPU & High-Performance Computing** I am partnering with an AI technology business to hire an infrastructure specialist who can keep demanding engineering workloads running reliably at scale. This is a hands-on role with substantial ownership. You will manage the compute environments used for development, benchmarking and testing, making accelerator capacity easier to access, failures faster to diagnose, and infrastructure more efficient to operate. We welcome backgrounds in **AI infrastructure, research computing, cloud platforms, SRE and performance-sensitive financial technology environments .** **Highlights:** * Work with a global elite team. * Equity participation considered on a case-by-case basis. * Comprehensive and competitive benefits package. **The opportunity** You will be the person engineers turn to when compute becomes a bottleneck. Your remit will cover Linux systems, accelerator-equipped servers, workload scheduling, connectivity and storage across externally hosted and dedicated environments. Key responsibilities include: * Keep **clusters operational** through proactive health checks, capacity management and preventative maintenance. * Investigate failures across hardware, drivers, networking, storage and scheduling, taking incidents through to resolution. * Build monitoring and alerts that distinguish actionable problems from noise. * Automate provisioning, configuration and recovery to reduce repetitive operational work. * Improve resource allocation and identify capacity lost to idle nodes, failed jobs or competing workloads. * Establish consistent access controls, troubleshooting procedures and operating standards. * Work with engineering stakeholders and compute providers to support changing workload and capacity needs. **What we are looking for** You have operated infrastructure that engineers or researchers depend on and are comfortable taking responsibility when it fails. Your experience should include: * **1 yrs+ -Administration of GPU clusters, HPC systems or comparable research and production computing environments.** * Strong Linux troubleshooting skills, including processes, permissions, networking, storage, package management and logs. * Practical experience diagnosing accelerator-server issues, such as driver incompatibilities, unhealthy devices, memory faults and node failures. * Workload orchestration or batch scheduling using Kubernetes, Slurm or similar systems. * Scripting and infrastructure automation with Python, Bash, configuration management or infrastructure-as-code tools. * Clear communication during incidents and the ability to work independently through unfamiliar technical problems. **Experience that would stand out** * Responsibility for accelerator fleets numbering in the thousands. * Operating across multiple hosting providers or different accelerator architectures. * Improving cluster utilisation or reducing recurring failures and recovery times. * High-speed interconnects, collective communications or distributed workload performance. * High-throughput storage and distributed filesystems. * Secure infrastructure access, identity management and secrets handling. * Building operational tooling, runbooks and repeatable deployment standards.