Senior Product Architect, Storage
$224k - $357k • Remote • US, CA, Santa Clara
Posted 1d ago
About the job
NVIDIA is seeking an AI Storage Platform Architect to bridge the gap between advanced hardware and AI deployments. This role involves translating the capabilities of NVIDIA's latest hardware, including Rubin GPUs, Vera CPUs, BlueField DPUs, NVLink, and Spectrum-X networking, into production-ready blueprints. You will collaborate with storage ecosystem partners to co-develop reference architectures for the NVIDIA AI Data Platform, optimizing every layer of the stack for AI workloads. This is an opportunity to work in a diverse and supportive environment, making a significant impact on the future of computing.
Responsibilities
- Architect end-to-end reference architectures for disaggregated inference, large-scale foundation model training, and agentic AI pipelines.
- Design and validate storage-optimized AI infrastructure, including KV Cache tiering, checkpoint acceleration, and high-throughput dataset pipelines.
- Define system-level architectures for GPUs, CPUs, DPUs, and networking to enhance AI lifecycle efficiency.
- Develop and publish reference architectures, whitepapers, and deployment guides.
- Drive prototyping, benchmarking, and performance validation of AI infrastructure at scale.
- Leverage DOCA to architect DPU-offloaded data services for storage acceleration, telemetry, security, and network virtualization.
- Collaborate with AI teams to build retrieval-optimized storage architectures, including vector database integration and low-latency access patterns.
- Partner with customers and collaborators to co-innovate and deliver proof-of-concepts.
Requirements
- 12+ years of experience architecting datacenter-scale AI, HPC, or storage infrastructure.
- Bachelor's degree in Computer Science or related field, or equivalent experience.
- Deep expertise in AI infrastructure, including disaggregated inference, LLM training pipelines, and autonomous AI systems.
- Hands-on experience with RDMA, high-performance storage protocols (NVMeoF, GPFS, Lustre, S3), and low-latency fabric design.
- Strong understanding of KV Cache management strategies and tiered memory/storage hierarchies.
- Familiarity with Retrieval-Augmented Generation (RAG) architectures and their storage/retrieval demands.
- Experience with NVIDIA DOCA or equivalent DPU/SmartNIC programming frameworks.
- Proven foundation in networking, including Ethernet, InfiniBand, NVLink, congestion control, and datacenter topologies.
Benefits
- Equity
- Benefits