Solution Architect (GPU/Compute)

Own the technical architecture for GPU and accelerated compute infrastructure, translating AI training, inference, HPC, capacity, performance, and scaling requirements into deployable compute solutions. This role is a customer-facing technical leader supporting the sales organization throughout discovery, solution development, technical validation, and transition to deployment. The architect works as part of the broader AI infrastructure architecture team and collaborates closely with the other specialist domains to deliver an integrated end-to-end solution.

What you'll be doing

• Lead technical discovery and architecture for GPU / Compute opportunities within strategic AI Data Center programs.
• Design scalable GPU clusters and accelerated compute platforms across GPUs, CPUs, memory, PCIe, NICs/DPUs, local NVMe, systems management, and supporting infrastructure.
• Translate workload, performance, capacity, resiliency, and growth requirements into platform sizing, configurations, reference architectures, BOMs, and technical standards.
• Evaluate and position current and next-generation GPU, server, and rack-scale compute platforms based on customer requirements.
• Define compute architecture dependencies across networking, storage, power, cooling, rack design, and physical deployment.
• Support cluster bring-up, firmware and software validation, benchmarking, performance optimization, troubleshooting, and production readiness.
• Partner with Network and Storage Solution Architects to ensure balanced end-to-end AI cluster performance.
• Work with Rack Integration, Fiber, Services Architecture, OEMs, Professional Services, and delivery teams to ensure designs can be integrated and deployed at scale.
• Support RFQs/RFPs, proposals, customer workshops, technical presentations, proofs of concept, and architecture reviews.
• Develop repeatable compute reference architectures and technical standards for large-scale AI Data Center programs.

What you have

• 7+ years of experience in data center compute, systems engineering, solution architecture, HPC, AI infrastructure, or related technical roles.
• Deep expertise in GPU / accelerated compute, server architecture, HPC, or large-scale compute infrastructure.
• Strong knowledge of GPU servers, CPU/GPU topology, memory, PCIe, NICs/DPUs, NVMe, firmware, BIOS/BMC, and systems management.
• Experience designing or supporting large GPU clusters, AI training/inference environments, HPC platforms, hyperscale infrastructure, or cloud compute environments.
• Understanding of high-density power, cooling, networking, and storage dependencies associated with AI compute platforms.
• Experience with Linux, cluster deployment, validation, performance analysis, and complex infrastructure troubleshooting.
• Strong customer-facing architecture, documentation, workshop, and presentation skills.
• Ability to work across customers, OEMs, sales, engineering, Professional Services, integration, and delivery teams.
• NVIDIA DGX, HGX, GB200/GB300 NVL or comparable accelerated compute platforms
• Large-scale GPU cluster design and deployment
• NICs, SuperNICs, DPUs, NVLink/NVSwitch and high-speed GPU interconnect technologies
• Kubernetes, Slurm, cluster orchestration, provisioning, observability, or infrastructure automation
• AI CSP, hyperscale, NCP / neocloud, HPC, or large cloud infrastructure environments
• Cluster benchmarking, burn-in, validation, production bring-up, and performance optimization
• Technical Architecture – Brings deep domain expertise and translates customer requirements into scalable, supportable solutions.
• AI Infrastructure Knowledge – Understands how GPU / Compute, Network, and Storage operate together as an integrated AI platform.
• Design for Scale – Creates standardized architectures that can support large, rapidly expanding AI Data Center programs.
• Customer Engagement – Leads discovery, technical workshops, architecture reviews, and solution discussions with customer stakeholders.
• Performance & Troubleshooting – Uses data and technical analysis to identify bottlenecks, validate designs, and resolve complex infrastructure issues.
• Cross-Functional Leadership – Connects customers, OEMs, sales, engineering, Professional Services, integration, and delivery teams.
• Design-to-Delivery Thinking – Ensures technical architectures can be sourced, integrated, deployed, validated, and operated at scale.
• Quality, scalability, and technical accuracy of solution architectures
• Speed from customer requirements to validated technical design
• Successful technical support of AI Data Center pipeline and strategic pursuits
• Accuracy of sizing, BOMs, reference designs, and technical requirements
• Successful integration with the other AI infrastructure architecture domains
• Reduction in design changes, deployment issues, and performance bottlenecks
• Successful transition from solution architecture into production deployment
• Customer and OEM confidence in the technical solution
• Development and adoption of repeatable reference architectures and standards