Solutions Architect - CPU and LPU
Core
Drive adoption of next-generation AI infrastructure across NVIDIA CPU platforms and LPU-based inference systems, focusing on full-stack AI efficiency from model architecture to workload partitioning.
Role type
Senior IC solutions architect (AI systems & heterogeneous computing)
Builds
Proof-of-concept development, software stack optimization, and deployable reference architectures for CPU/GPU/LPU environments
Domain
AI infrastructure, LLM inference, heterogeneous computing (CPU/GPU/LPU)
Deliverable
production ML models
Required skills
Modern CPU architecture, Linux systems, software performance tuning, AI inference for LLM/generative AI, heterogeneous system optimization, PyTorch, Triton, TensorRT-LLM, vLLM, ONNX Runtime
Preferred skills
NVIDIA CPU platforms (Grace, Vera, Arm64), LPU-based systems, LLM inference optimization, customer-facing proof-of-concepts
Technologies
Grace, Vera, LPX-class systems, PyTorch, Triton, TensorRT-LLM, vLLM, ONNX Runtime
Responsibilities
Evangelize NVIDIA CPU and LPU platforms, design and optimize AI workloads across CPU/GPU/LPU, analyze and tune LLM/generative AI pipelines, build proof-of-concepts and reference architectures, establish trusted technical relationships with customer architects and leaders
Seniority
Senior, hands-on IC
