HPC Network Solutions Architect: Low-Latency, Multi‑Vendor

NorthMark Compute & Cloud; Dallas (TX); On-site; USD 140,000 - 190,000; Full time

Benefits offered by this job

Lunch stipend

Employer-paid medical benefits

Dental & Vision benefits

Paid parental leave

401(k) matching

Job summary

NorthMark Compute & Cloud is seeking an HPC Network Solutions Architect to design, integrate, and optimize high-performance networking for HPC, AI/ML, and data-intensive workloads.

You will guide customers through requirements, design, POC, deployment, and long-term optimization, aligning low-latency networking with workload needs and collaborating with internal teams to drive roadmap decisions.

Qualifications

• Experience in HPC networking solution architecture, systems design, or data center network engineering.
• Strong expertise in InfiniBand and RoCE protocols, including deployment and tuning at scale.
• Hands-on experience designing and implementing large-scale Ethernet networks (BGP, OSPF, EVPN, VXLAN).
• Deep understanding of GPU communication frameworks such as MPI and NCCL and their integration with HPC interconnects.
• Proficiency with Linux-based environments and scripting (Python, Bash, PowerShell) for automation.
• Experience in multi-vendor environments and evaluating new networking platforms.
• Ability to translate complex networking requirements into clear architectures and present to customers.
• Strong customer-facing communication skills for executives and technical stakeholders.

Responsibilities

• Act as the primary networking SME for customers adopting or scaling HPC environments.
• Capture network performance goals, scalability requirements, and integration constraints with customers.
• Design and document end-to-end HPC network architectures including Ethernet, InfiniBand, RoCE, EVPN, and VXLAN fabrics.
• Lead proof-of-concept and benchmarking engagements to validate low-latency/high-throughput designs.
• Optimize multi-vendor data center and HPC interconnects addressing scaling challenges like data gravity and throughput bottlenecks.
• Define integration strategies across compute, storage, orchestration, and security layers.
• Conduct network performance assessments and tuning, identifying bottlenecks and improvements.
• Build observability frameworks for HPC networks using Prometheus, Grafana, and vendor telemetry.
• Collaborate with engineering, product, and operations to refine architecture blueprints.
• Partner with ecosystem vendors to integrate features and influence roadmap evolution.
• Stay current with HPC networking technologies and provide future adoption insights.

Skills

HPC networking

InfiniBand

RoCE

Ethernet networks

BGP/OSPF

EVPN/VXLAN

MPI/NCCL

Linux scripting

Multi-vendor support

Customer-facing communication

Education

Bachelor’s or Master’s degree in Computer Science/Networking/Engineering

Cisco CCNP/CCIE, Juniper JNCIP, AWS Advanced Networking Specialty, Red Hat RHCE

Tools

Prometheus

Grafana

Telemetry tooling

Job description

NorthMark Compute & Cloud is seeking an HPC Network Solutions Architect to design, integrate, and optimize high-performance networking for HPC, AI/ML, and data-intensive workloads.

You will guide customers through requirements, design, POC, deployment, and long-term optimization, aligning low-latency networking with workload needs and collaborating with internal teams to drive roadmap decisions.