AI System Architect (NPU / GPU / Interconnect)
Core
Define next-generation AI and data center compute platforms by engaging with customers to translate requirements into scalable system architecture, technology roadmaps, and platform features.
Role type
Senior IC AI System Architect (NPU/GPU/Interconnect)
Builds
Next-generation AI and data center compute platforms
Domain
Semiconductor / AI Hardware / Data Center Infrastructure
Deliverable
production ML models
Required skills
AI accelerator/NPU architecture, GPU architecture, high-performance interconnects, HW/SW co-design, system-level architecture, memory hierarchy optimization, AI workload modeling
Preferred skills
LLM/Transformer optimization, chiplet communication, distributed AI systems, PCIe/CXL/Ethernet/UALink fabrics
Technologies
NPU, GPU, PCIe, CXL, Ethernet, UALink, LLMs, Transformers
Responsibilities
Engage with customers and stakeholders to understand AI workload requirements and system constraints; Translate requirements into scalable system architecture and technology roadmaps; Drive architecture tradeoff analysis across performance, power, cost, and scalability; Analyze and optimize AI workloads including LLMs and computer vision; Drive HW/SW co-design across architecture, compiler, runtime, and driver layers; Architect high-performance data movement and cache-coherent fabrics for distributed AI systems.
