GPU Architect
Core
Architect and design multi-GPU scale-up and scale-out systems for next-generation AI datacenter platforms, tightly coupling GPU compute, high-bandwidth memory, and interconnects.
Role type
Senior IC GPU System/Fabrics Architect
Builds
Next-generation AI datacenter platforms with hundreds/thousands of interconnected GPUs
Domain
Semiconductor hardware design / AI Infrastructure
Deliverable
production ML models
Required skills
Multi-GPU system architecture, High-speed interconnect protocols (NVLink, Ethernet, InfiniBand, CXL, PCIe), System modeling and bottleneck analysis, Hardware-software co-design, Multi-chiplet and multi-node rack-scale system design
Preferred skills
NIC/DPU/RDMA/RoCE architecture, Chiplet interconnect architectures
Technologies
NVLink, Ethernet, InfiniBand, CXL, PCIe, RDMA, RoCE
Responsibilities
Architect multi-GPU systems for scale-up and scale-out configurations; Define and evaluate future architectures for high-speed interconnects; Architect RDMA-capable hardware and define transport layer optimizations; Explore and build novel high-density multi-chiplet, multi-package, multi-node rack-scale AI systems; Use system models and simulations to guide design trade-offs; Collaborate with GPU ASIC, compiler, library, and software teams.
Seniority
Senior, hands-on IC
