System Engineer

Hermes Corporate; Brasil; On-site; BRL 300,000 - 420,000; Full time

Job summary

Hermes Corporate is seeking an experienced AI & HPC Infrastructure Engineer to operate, optimise, and evolve high-performance computing infrastructure for engineering and scientific workloads.

This hands-on role requires deep Linux and HPC expertise, owning production platforms, troubleshooting complex systems, and driving improvements across GPU compute, storage, networking, containers, and platform performance.

Qualifications

• Extensive Linux experience at an advanced level.
• Proven HPC production environment ownership and cluster management.
• Hands-on GPU compute infrastructure with NVIDIA stack.

Responsibilities

• Administer GPU compute infrastructure, including drivers, firmware, and health monitoring.
• Operate and optimise HPC clusters for reliability and performance.
• Define resource allocation policies for diverse workloads.
• Manage containerised environments for platform users with GPU access.
• Maintain high-speed storage and interconnects operation and troubleshooting.
• Lead performance engineering: profiling, benchmarking, bottleneck analysis.
• Support integration of applications on GPU-accelerated HPC platforms.
• Provide incident response, root-cause analysis, and change management.
• Maintain platform monitoring, alerting, and documentation.
• Contribute to security, architecture governance, and compliance.

Skills

Linux systems engineering

HPC cluster administration

GPU compute infrastructure

Containerisation

Performance troubleshooting

End-to-end production ownership

English proficiency

Tools

NVIDIA drivers/SDK

CUDA toolkit

Container runtimes (Docker/Singularity)

Job description

We are looking for an experienced AI & HPC Infrastructure Engineer to operate, optimise, and continuously evolve high-performance computing infrastructure supporting advanced engineering and scientific workloads.

This is a hands-on infrastructure role for someone with strong Linux and HPC expertise who is comfortable owning production platforms, troubleshooting complex systems, and driving improvements across GPU compute, storage, networking, containers, and platform performance.

What You’ll Do

• Administer and maintain GPU compute infrastructure, including system configuration, NVIDIA drivers and software stack, firmware, and hardware health monitoring.
• Operate, monitor, and optimise HPC clusters to ensure reliability, performance, and efficient use of compute resources.
• Define and refine resource allocation policies across different workload types to ensure effective and equitable use of available capacity.
• Manage containerised environments for platform users, including user provisioning, environment maintenance, GPU access, and standards for container usage.
• Administer shared and high-performance storage and support the operation and troubleshooting of high-speed interconnects.
• Lead performance engineering activities, including profiling, benchmarking, bottleneck analysis, and platform optimisation.
• Support the integration and optimisation of engineering and scientific applications on GPU-accelerated and HPC platforms.
• Provide incident response, troubleshooting, root-cause analysis, and change management, including planning and execution of maintenance activities.
• Maintain platform monitoring, alerting, and operational reporting.
• Develop and maintain technical documentation, operational procedures, and platform standards.
• Contribute to security, architecture governance, and compliance activities.

Required Qualifications

• Substantial professional experience in Linux systems engineering / administration at an advanced level.
• Proven experience operating production HPC environments, including cluster administration and large-scale parallel workloads.
• Hands-on experience administering GPU compute infrastructure, particularly NVIDIA-based environments and the associated software stack.
• Practical experience with containerisation in HPC environments, including GPU device access and multi-node workloads.
• Strong troubleshooting and performance-tuning skills across complex compute infrastructure.
• Demonstrated ability to take end-to-end ownership of production infrastructure and act as a senior technical escalation point.
• Strong analytical and problem-solving skills with a production-focused mindset.
• Professional working proficiency in English.

Preferred Qualifications

• Experience supporting engineering simulation, scientific computing, or computational analysis workloadson accelerated infrastructure.
• Experience with HPC scheduling and resource management.
• Experience with monitoring and observability for compute infrastructure.
• Experience with high-performance storage and networking/interconnect technologies.
• Experience in an enterprise, industrial, or regulated environment.
• Experience with platform automation, scripting, or infrastructure tooling.

Ideal Candidate

The ideal candidate is a hands-on infrastructure engineer who combines deep Linux, HPC, and GPU expertise with strong operational ownership.

You are comfortable working close to the hardware and software stack, diagnosing complex performance and reliability issues, and improving infrastructure used by demanding engineering and scientific workloads. You thrive in environments where reliability, performance, and technical depth matter, and you can operate effectively as a senior technical point of reference for the platform.