Senior AI Systems Architect — Hybrid HW/SW Co‑Design

SanDisk; Milpitas (CA); Hybrid; USD 180,000 - 306,000; Full time

Job summary

Sandisk seeks a Senior AI Systems Architect in Milpitas, CA, to lead architecture for AI inference accelerators with HBF technology. You will design SW-HW co-design, build performance models, and perform competitive analysis for internal and customer-facing specs.

The role covers interfaces, compute/memory configuration, security, telemetry, reliability, power and performance requirements, with a strong emphasis on AI workflows and system-level integration.

Qualifications

• PhD or MS in Electrical & Computer Engineering or Computer Science with 15 years industry experience in system-level architecture.
• Deep understanding of I/O and memory interfaces including DDR, HBM, PCIe, CXL, NVLink, and related interconnects.
• Experience with AI inference accelerators from leading vendors is a plus.
• Knowledge of AI inference frameworks and KV Cache management, RAG, Agentic AI trends is valuable.
• Strong programming skills and ability to generate models for simulating architectures.

Responsibilities

• Conduct pathfinding and develop architecture for AI Inference Accelerators leveraging Sandisk's HBF Technology.
• Define HW/SW interfaces, compute/memory architecture, and configuration requirements for security, telemetry, reliability, power and performance.
• Contribute to PoC development and write detailed architecture specifications for internal and external teams.

Skills

System architecture

I/O interfaces

DDR memory

PCIe CXL NVLink

AI inference

Programming skills

Modeling & simulation

Education

PhD or MS in Electrical & Computer Engineering or Computer Science

Tools

NVIDIA GPUs

AMD GPUs

DDR technologies

HBM memory

PCIe

CXL

NVLink

RDMA

NVMe

Job description

Sandisk seeks a Senior AI Systems Architect in Milpitas, CA, to lead architecture for AI inference accelerators with HBF technology. You will design SW-HW co-design, build performance models, and perform competitive analysis for internal and customer-facing specs.

The role covers interfaces, compute/memory configuration, security, telemetry, reliability, power and performance requirements, with a strong emphasis on AI workflows and system-level integration.