Principal System Cloud Architect

We are seeking a Principal System Architect to lead the architectural vision for NVIDIA’s GPU Infrastructure-as-a-Service (IaaS) offerings. This strategic role focuses on defining reference architectures and system blueprints that integrate NVIDIA’s latest innovations — including GB200 Grace Blackwell systems, Spectrum-X, Bluefield, InfiniBand, Storage (Block, File, Object), and AI Enterprise software stacks — into scalable, high-performance cloud infrastructure for on-prem, Neo Clouds, and CSPs.

This role requires deep engagements across hardware, networking, orchestration, and partner ecosystems to define the future of GPU cloud services.

What you’ll be doing:

• Architect Future-Ready GPU Infrastructure

• Define scalable, secure, and efficient architectures for GPU-based IaaS using NVIDIA’s full stack: DGX/HGX, GB200, NVLink/NVSwitch, InfiniBand, and Spectrum-X.

• Lead Reference Architecture Development

• Work with internal engineering, cloud partners, and OEMs to define and publish validated reference architectures covering bare-metal provisioning, virtualization, storage fabrics, and networking.

• Drive End-to-End Cloud Infrastructure Strategy

• Architect solutions for bare-metal-as-a-service, VMaaS, and container orchestration (Kubernetes), integrated with virtual networking (VPCs), Infiniband fabrics, high-performance storage, and AI workloads.

• Influence Product Strategy Across Domains

• Partner with silicon, platform, networking, and software teams to ensure alignment of architecture with NVIDIA’s roadmap for GPU, DPU, and AI services.

• Engage with Ecosystem Partners

• Represent NVIDIA in joint solution development with CSPs, OEMs, and hyperscale customers to align infrastructure strategies and deployment practices.

• Evaluate Trade-Offs and Drive Decisions

• Make high-impact architectural decisions across performance, scalability, multi-tenancy, power efficiency, and manageability

What we need to see:

• 15+ years in system architecture, with deep experience in cloud-scale infrastructure, HPC, or AI platforms.

• Proven expertise in GPU platforms, data center networking (InfiniBand, RoCE, Spectrum), virtual networking, storage, and orchestration technologies.

• Strong understanding of Kubernetes, VM provisioning, bare-metal provisioning, and infrastructure automation.

• MS or PhD in Computer Engineering, Electrical Engineering, related field, or equivalent experience..

• Demonstrated ability to define, document, and present architectural designs and influence cross-functional teams.

Ways to stand out from the crowd:

• Experience with NVIDIA technologies such as DGX, HGX, GB200, NVLink, NVSwitch, BlueField, Magnum IO, and Spectrum-X.

• Deep knowledge of AI/ML workloads, distributed training architectures, and GPU scheduler integration.

• Familiarity with CSP environments (AWS, Azure, OCI, GCP) and hybrid/multi-cloud architectures.

• Participation in open standards and industry bodies (e.g., OCP, CNCF, Kubernetes SIGs).

With competitive salaries and a generous benefits package, NVIDIA is widely considered to be one of the technology world’s most desirable employers; we have some of the most forward-thinking and hardworking people in the world working for us and, due to unparalleled growth, our outstanding teams are rapidly growing.