Senior Infrastructure Engineer, Research (AI/HPC)

Location: Singapore; hybrid model with office time and work-from-home days.

Company

hirify.global is a physics AI company building simulation software for industrial engineering and manufacturing, working with aerospace and defence, automotive, semiconductor, materials, and energy and renewables organisations.

What you will do

• Design, build, maintain, harden, and optimize SLURM-on-Kubernetes infrastructure for large-scale distributed training and data-generation workloads.
• Own cluster reliability, efficiency, and utilization across hundreds of GPUs, and contribute to capacity planning and the infrastructure roadmap.
• Partner with Research to debug job failures, diagnose performance bottlenecks, and improve iteration speed.
• Build and maintain reproducible containerized research environments using Docker, Enroot, and Pyxis.
• Manage high-performance storage for data-intensive CAE workloads.
• Develop tooling, automation, observability, and a technical scaling strategy for efficient large-scale model training and data generation.

Requirements

• 4–6 years of experience in infrastructure, platform, or HPC engineering roles.
• Hands-on experience with SLURM and Kubernetes, ideally running them together in production environments.
• Experience with containerization and orchestration for compute-heavy workloads, including Docker and Enroot/Pyxis or similar tools.
• Working knowledge of high-performance networking, including InfiniBand, RDMA, and NCCL, in distributed-training contexts.
• Experience managing HPC/ML storage systems at scale.
• Strong debugging skills across scheduling, networking, and individual job failures.
• Ability to work directly with researchers and translate their needs into infrastructure improvements.
• Solid scripting and automation skills.

Nice to have

• Experience operating GPU clusters at scale, with hundreds of GPUs or more.
• Background in CAE or simulation-driven data generation.
• Experience with GPU-cluster health monitoring and fault tolerance for long-running training jobs.
• Familiarity with infrastructure-as-code tools such as Terraform and Helm.

Culture & Benefits

• Opportunity to shape an AI-native engineering company at a formative stage and work on real-world industrial and societal challenges.
• Collaborative environment with engineers, scientists, and operators from diverse backgrounds.
• Flat structure where ideas are valued regardless of hierarchy, and challenging assumptions is encouraged.
• Hybrid work model combining time in the Shoreditch office with work-from-home days.
• Commitment to equal employment opportunity and support for underrepresented groups in technology.