Principal Solutions Architect - AI/HPC

Cirrascale; San Diego (CA); Hybrid; USD 210,000 - 250,000; Full time

Job summary

Cirrascale is seeking a Principal Solutions Architect - AI/HPC to bridge Sales Engineering, Engineering Infrastructure, and Data Center Operations. You will lead technical engagements with high-profile customers to design scalable GPU clusters for training and inference, turning complex requirements into deployable architectures and BOMs.

You will balance customer needs with physical facility constraints, applying systems thinking to deliver high-density power, cooling solutions, and low-latency

Qualifications

• 5+ years of experience in Solutions Architecture or Field Applications Engineering focused on AI, HPC and networking.
• Hands-on experience designing, provisioning, or operating multi-node GPU clusters at a scale of 128 or more.
• Comprehensive understanding of high-speed interconnects (InfiniBand, RDMA, RoCEv2) and data center layouts.
• Proficiency in Linux administration, container platforms, and cluster scheduling tools.

Responsibilities

• Design high-performance AI/HPC solutions aligning customer goals with data center constraints.
• Translate OEM reference architectures into physical infrastructure, accounting for cooling and power delivery.
• Create architecture diagrams, rack layouts and BOMs to guide deployments.
• Collaborate with Data Center Operations on rack placement and cooling.
• Develop low-latency networks and HPC storage to maximize GPU performance.
• Build, test and demonstrate PoCs to validate proposed hardware solutions.

Skills

AI & HPC architecture

Linux administration

Networking concepts

Technical diagramming

Tools

InfiniBand

RoCEv2

NVIDIA Hopper/Blackwell

GPFS/Lustre

WEKA storage

Job description

Cirrascale Cloud Services provides high-performance cloud infrastructure purpose-built for deep learning, generative AI, and large-scale AI inference workloads. We specialize in dedicated GPU cloud solutions tailored to the unique needs of startups, research labs, and enterprise AI teams. Our mission is to accelerate AI innovation by combining powerful hardware with white-glove service and flexible, custom-built environments.

Position Summary

We are seeking a hands-on Principal Solutions Architect - AI/HPC to serve as our primary technical solutions bridge across Sales Engineering, Engineering Infrastructure, and Data Center Operations. This role will report directly to the VP of Engineering.

In this high-impact role, you will lead the technical engagement with high-profile customers and Sales Engineering to design high-scale GPU clusters tailored for training and inference workloads. You will apply a strong systems engineering mindset to balance customer requirements with physical facility constraints, successfully turning complex requirements into architecture diagrams and a deployable Bill of Materials.

Key Responsibilities

• Design high-performance AI/HPC solutions that align with customer goals, partnering closely with Sales Engineering to bridge the gap between technical capability, Data Center constraints and customer needs.
• Translate OEM reference architectures into physical infrastructure solutions; proactively accounting for extreme high-density power delivery, liquid/air cooling demands, and ultra-low-latency backend fabrics.
• Translate requirements into architecture diagrams, rack diagrams and BOMs.
• Partner with Data Center Operations to evaluate rack placement, structural floor load limits, power distribution and cooling.
• Design low-latency, lossless backend networks (InfiniBand, RoCEv2, and TCP/IP).
• Design HPC storage systems (such as WEKA) to maximize GPU performance.
• Build, test and demonstrate PoCs to showcase proposed hardware solutions.

Requirements

• 5+ years of experience in Solutions Architecture or Field Application Engineering focused on AI, HPC and networking.
• Practical understanding of data center physical constraints, including 3-phase power distribution, rack power shelves, airflow CFM requirements, and Direct-to-Chip liquid cooling systems.
• Experience designing high-density power limits, advanced water-and-air cooling dynamics, and rack-level layouts.
• Hands-on experience designing, provisioning, or operating multi-node GPU clusters at a scale of 128 or more.
• Comprehensive understanding of high-speed interconnects (InfiniBand, RDMA, RoCEv2), high-radix Ethernet switches, and copper/optical media types (DAC, AOC, SMF/MMF).
• Advanced knowledge of fast, parallel distributed storage platforms (WEKA, GPFS, Lustre).
• Deep familiarity with NVIDIA Hopper and Blackwell architecture and associated software stacks.
• Proficiency in Linux administration, network operating systems, container platforms, and cluster scheduling tools
• Strong written and oral English communication skills, with a proven ability to illustrate complex architectures in diagrams and explain technical trade-offs to executive, sales, and operations teams.
• Strong time-management, organization, and problem-solving skills to manage multiple concurrent projects and navigate customer infrastructure challenges independently.

Salary Range

The base salary range for the Principal Solutins Architect - AI/HPC is $210,000 to $250,000 USD. This pay range reflects the broad, minimum to maximum, pay range for this job for the location for which it has been posted. Compensation decisions are dependent on several factors including, but not limited to, an individual's qualifications, location where the role is to be performed, internal equity, and alignment with market data.

Comprehensive benefits package, including health, dental, and vision insurance, retirement plans, paid time off, and opportunities for professional development.

Why Join Cirrascale?

Join a growing team that's pushing the boundaries of AI infrastructure. At Cirrascale, you'll contribute to projects powering next-generation AI applications while working with top-tier hardware in a collaborative and innovative environment. From custom deployments to hands-on customer support, every role here plays a part in enabling breakthroughs in AI.