AI Infrastructure Engineer

Cypress Consulting; Washington (District of Columbia); Hybrid; USD 413,000 - 482,000; Full time

Job summary

Cypress Consulting is seeking an AI Infrastructure Engineer to bring rack-scale AI systems online and prepare them for reliable operation. This onsite role in Washington, DC focuses on configuring compute, networking, and software platforms within a restricted-network environment.

The candidate will deploy and validate accelerator software, integrate high-performance networking, and coordinate with infrastructure teams to ensure smooth handoff and acceptance. U.S.

Qualifications

• Hands-on experience deploying rack-scale AI or HPC infrastructure (e.g., NVIDIA GB200/GB300, AMD Instinct clusters)
• Experience configuring high-performance AI networking and accelerator interconnects (e.g., Spectrum-X, InfiniBand)
• Working knowledge of Linux-based accelerator software and cluster management platforms
• Experience deploying systems in disconnected or restricted-network environments
• Ability to plan and execute infrastructure validation and clearly document results
• U.S. citizenship and ability to complete site access and background screening requirements

Responsibilities

• Configure and bring online rack-scale AI compute systems and their supporting interconnects.
• Integrate high-performance cluster networking and connectivity with the surrounding infrastructure.
• Deploy and configure cluster management, provisioning, monitoring, and scheduling capabilities.
• Install and maintain Linux, accelerator software, and related components using appropriate offline deployment practices.
• Run functional and performance validation, resolve configuration issues, and document operational readiness.
• Coordinate with infrastructure and network teams during system handoff and acceptance.

Skills

Rack-scale AI infra

AI networking

Linux software

Restricted-network deployment

Validation documentation

Background screening

Job description

Onsite, Washington, DC Onsite full time Full-time $325/hour Project: 3 months

Role overview

We are seeking an AI Infrastructure Engineer to bring rack-scale AI systems online and help prepare them for reliable operation.

This project focuses on the logical configuration, integration, and validation of compute, networking, and software platforms in a restricted-network environment.

Responsibilities

• Configure and bring online rack-scale AI compute systems and their supporting interconnects.
• Integrate high-performance cluster networking and connectivity with the surrounding infrastructure.
• Deploy and configure cluster management, provisioning, monitoring, and scheduling capabilities.
• Install and maintain Linux, accelerator software, and related components using appropriate offline deployment practices.
• Run functional and performance validation, resolve configuration issues, and document operational readiness.
• Coordinate with infrastructure and network teams during system handoff and acceptance.

Required qualifications

• Hands-on experience deploying rack-scale AI or high-performance computing infrastructure, such as NVIDIA GB200/GB300, AMD Instinct clusters, or equivalent platforms.
• Experience configuring high-performance AI networking and accelerator interconnects, such as NVIDIA Spectrum-X, Quantum InfiniBand, or comparable fabrics.
• Working knowledge of Linux-based accelerator software and cluster management platforms.
• Experience deploying systems in disconnected or restricted-network environments.
• Ability to plan and execute infrastructure validation and clearly document results.
• U.S. citizenship and ability to complete site access and background screening requirements.

Preferred qualifications

• Familiarity with NVLink/NVSwitch or RoCEv2 and cluster orchestration using Slurm or Kubernetes.
• Experience with CUDA or ROCm software stacks and validation tools such as NCCL/RCCL.
• Relevant AI infrastructure or networking certifications, such as NVIDIA-Certified Professional (AI Infrastructure or Networking), Arista ACE:L5, Cisco CCIE Data Center, or comparable credentials.
• Prior experience deploying systems in secure or restricted-access facilities.