HPC Systems Architect

Autonomai Recruitment; Chicago (IL); On-site; USD 150,000 - 190,000; Full time

Job summary

Autonomai Recruitment seeks an HPC Systems Architect for Chicago's research infrastructure. You will shape the compute, storage, and networking stack for high-demand ML and quantitative research workloads, operating across global environments.

The role is hands-on, requiring experience from design through delivery, with attention to performance, scalability, and reliability. Collaboration with researchers and engineers is essential to translate workload needs into practical infrastructure

Qualifications

• Extensive Linux HPC engineering and architecture experience.
• Experience with workload scheduling and orchestration.
• Familiarity with high-performance storage technologies.
• Background in GPU infrastructure for AI/ML workloads.
• Strong networking knowledge for fast data movement.

Responsibilities

• Design and guide HPC infrastructure for research workloads.
• Evaluate CPU/GPU compute, storage, scheduling, and networking.
• Diagnose bottlenecks and improve throughput and scalability.
• Collaborate with researchers and engineers to translate requirements.
• Develop monitoring and analytics for performance visibility.

Skills

Linux HPC

Slurm/Kubernetes

High-performance storage

GPU/AI infrastructure

Networking (InfiniBand, high-speed)

Datacentre design

Automation languages (Python/Go/C++)

Education

BS or MS in Computer Science / Electrical Engineering

Tools

Slurm

Kubernetes

DDN/VAST/Isilon storage

GPU compute frameworks

Job description

HPC Systems Architect | Research Infrastructure

Chicago

The speed of research depends on the systems behind it.

We’re working with a research-driven trading firm to hire an HPC Systems Architect who will shape the infrastructure supporting quantitative research and machine learning across its global business.

This is a hands-on architecture role for someone who has built and operated large-scale compute environments. You’ll make decisions across compute, storage, scheduling and networking, then work with engineering teams to turn those designs into reliable production systems.

The challenge is to make the whole environment perform: keeping processors fed, moving data efficiently, allocating resources intelligently and finding the bottlenecks that slow researchers down.

What you’ll own

• Design and guide the implementation of HPC infrastructure for demanding research and machine learning workloads.
• Evaluate hardware and software across CPU and GPU compute, storage, scheduling and high-speed networking.
• Diagnose performance bottlenecks and improve throughput, utilisation, scalability and reliability.
• Establish architectural standards that support consistent expansion across global environments.
• Work directly with researchers and software engineers to translate workload requirements into practical infrastructure decisions.
• Develop monitoring and analytics that make system performance and capacity constraints visible.
• Shape policies for allocating compute, storage and networking resources.
• Take projects from initial design through delivery, including technical decisions, budgets and timelines.
• Advise senior leadership on technologies that could materially improve research capability.

What you’ll bring

You’ve spent around 10+ years building, operating and designing substantial compute environments, with the depth to explain why an architecture works and where it will fail.

Your experience should cover several of the following:

• Linux systems engineering and HPC architecture.
• Workload scheduling and orchestration, including Slurm, Kubernetes or comparable platforms.
• High-performance storage, including technologies such as DDN, VAST or Isilon.
• GPU infrastructure and the integration of AI/ML workloads.
• InfiniBand, high-speed Ethernet and network performance.
• Datacentre design, capacity planning and infrastructure expansion.
• Python, Go, C++ or similar languages for automation, tooling and systems development.

You’ll also need the communication skills to explain technical trade-offs, influence decisions and coordinate delivery across teams. A degree in computer science, electrical engineering or a related discipline is expected.

Why consider it?

You’ll have architectural influence across a global HPC environment, direct access to the people using it and responsibility for seeing your decisions through to production.

We’re particularly interested in systems engineers and HPC engineers who have grown into architecture through building and operating infrastructure.

Get in touch with a brief overview of the environments you’ve built, the scale involved and the architectural decisions you’ve owned.