Principal Core Infrastructure Engineer – GPU

Oracle; Hong Kong; Sur place; HKD 900 000 - 1 200 000; Plein temps

Résumé du poste

Oracle is seeking a Principal GPU/HPC Infrastructure Engineer to help power AI and HPC workloads in OCI. You will deploy and optimize multi-node GPU clusters, manage distributed systems, and own infrastructure automation while engaging directly with customers on deployment and production operations.

The role focuses on GPU infrastructure, cloud technologies, and high-performance networking, with exposure to Slurm, Kubernetes, and related tools. Strong Linux and scripting skills are essential.

Qualifications

• Hands-on experience deploying, managing, or troubleshooting multi-node GPU clusters or GPU-enabled HPC environments.
• Strong Linux systems engineering with production operations and optimization skills.
• Experience with GPU/HPC workload orchestration using Slurm, Kubernetes, PBS, or equivalent.
• Programming or scripting in Python or shell for automation and tooling.

Responsabilités

• Deploy, configure, manage, and troubleshoot multi-node GPU clusters and Linux HPC environments for AI/ML workloads.
• Manage distributed computing infrastructure including GPUs, networking, storage, and cluster management.
• Configure and optimize workload scheduling/orchestration with Slurm, Kubernetes, or equivalents.
• Develop automation scripts to streamline provisioning, configuration, and monitoring.
• Collaborate with customers and Oracle teams to resolve complex infra issues and improve outcomes.

Connaissances

GPU cluster deployment

Linux systems engineering

GPU/HPC workload orchestration

Python scripting

Networking & distributed systems

Customer-facing experience

Analytical & communication skills

Outils

Slurm

Kubernetes

PBS

Terraform

Ansible

Prometheus

Description du poste

Principal Core Infrastructure Engineer – GPU/HPC & AI Infrastructure

About the Role

Help power the next generation of AI and high-performance computing with Oracle Cloud Infrastructure (OCI).

We are seeking a highly skilled, hands-on Principal GPU/HPC Infrastructure Engineer to help strategic customers deploy, manage, and optimize large-scale GPU computing environments supporting AI and machine learning workloads.

In this customer-facing engineering role, you will work at the intersection of GPU infrastructure, high-performance computing (HPC), cloud technologies, and distributed systems. You will collaborate directly with customer engineering teams and Oracle's cloud engineering organization to solve complex infrastructure challenges, from initial proof-of-concept (POC) deployments to production-scale operations.

You will take technical ownership of GPU cluster deployment, workload orchestration, infrastructure automation, performance optimization, and troubleshooting, helping customers build scalable, reliable, and efficient AI infrastructure.

This opportunity is ideal for engineers who enjoy hands-on technical problem-solving, working with advanced GPU technologies, and engaging directly with customers on complex infrastructure deployments.

Key Responsibilities

• Deploy, configure, manage, and troubleshoot multi-node GPU clusters and Linux-based HPC environments supporting AI/ML workloads.
• Manage distributed computing infrastructure, including GPU compute nodes, high-performance networking, storage, and cluster management systems.
• Configure and optimize GPU workload scheduling and orchestration using technologies such as Slurm, Kubernetes, or equivalent platforms.
• Diagnose and resolve complex infrastructure issues involving GPU performance, networking, resource utilization, scalability, and system reliability.
• Identify infrastructure bottlenecks and implement performance tuning and optimization solutions.

Infrastructure Automation & Operations

• Develop and maintain automation scripts using Python, shell scripting, or similar technologies to streamline cluster provisioning, configuration, monitoring, and operations.
• Implement infrastructure automation, observability, and operational improvements to enhance performance and reliability.
• Maintain secure, scalable, and resilient infrastructure supporting demanding AI training and inference workloads.
• Document infrastructure configurations, technical solutions, troubleshooting procedures, and operational best practices.

Customer-Facing Technical Engineering

• Work directly with OCI strategic customers during technical evaluations, POCs, production deployments, and ongoing infrastructure operations.
• Collaborate with customer engineering teams to understand technical requirements, troubleshoot complex issues, and optimize GPU/HPC environments.
• Take ownership of critical customer infrastructure challenges, conduct root-cause analysis, and drive effective technical resolutions.
• Provide guidance on GPU infrastructure architecture, performance optimization, and migration of AI/HPC workloads to the cloud.
• Partner with Oracle's core engineering, product, and support teams to resolve complex technical issues and improve customer outcomes.

Required Qualifications

• Demonstrated hands-on experience deploying, managing, or troubleshooting multi-node GPU clusters or GPU-enabled HPC environments, with practical responsibility for cluster infrastructure.
• Strong Linux systems engineering expertise, including advanced troubleshooting, system configuration, performance optimization, and production operations.
• Practical experience with GPU/HPC workload orchestration or cluster management technologies such as Slurm, Kubernetes, PBS, or equivalent platforms.
• Hands-on experience with Python, shell scripting, or similar technologies for infrastructure automation, configuration, and troubleshooting.
• Solid understanding of distributed computing, networking, storage, and infrastructure performance optimization.
• Proven ability to investigate and resolve complex infrastructure performance, scalability, and reliability issues.
• Experience working directly with external customers or customer technical teams on infrastructure deployment, technical troubleshooting, or production support.
• Strong analytical, communication, and collaboration skills, with the ability to independently drive technical problem resolution.

Preferred Qualifications

• Hands-on experience with NVIDIA GPU technologies, including A100, H100, H200, B200, GB200, GB300, or similar accelerator platforms.
• Experience with high-performance networking technologies such as InfiniBand, RDMA, RoCE, or NVIDIA/Mellanox networking solutions.
• Familiarity with GPU interconnect and distributed computing technologies such as NCCL, NVLink, or multi-node distributed training environments.
• Experience operating large-scale GPU clusters within cloud providers, GPU neocloud platforms, HPC facilities, or enterprise data centers.
• Experience with public cloud platforms such as OCI, AWS, Azure, Google Cloud, Alibaba Cloud, Tencent Cloud, or equivalent infrastructure environments.
• Familiarity with infrastructure-as-code, configuration management, and monitoring tools such as Terraform, Ansible, Prometheus, or similar technologies.
• Understanding of AI/ML infrastructure requirements and frameworks such as PyTorch or TensorFlow.

Why Join Oracle?

Build and optimize the infrastructure powering the AI revolution.

At Oracle Cloud Infrastructure, you will help strategic customers tackle challenging GPU computing and high-performance infrastructure problems.

• Work with advanced GPU infrastructure: Support demanding AI training, inference, and distributed computing workloads.
• Solve complex engineering challenges: Take hands-on ownership of GPU clusters, high-performance networking, workload orchestration, automation, and performance optimization.
• Deliver direct customer impact: Help strategic customers successfully deploy and operate AI infrastructure from POC through production.
• Collaborate with cloud engineering experts: Work closely with Oracle's core infrastructure, engineering, and product teams.
• Grow your technical expertise: Deepen your experience in GPU computing, HPC, distributed systems, and next-generation cloud infrastructure.

If you are passionate about GPU infrastructure, enjoy solving complex technical problems, and want to help customers scale their AI initiatives, we would like to hear from you.