Senior AI Inference Architect for Multi-Node GPU Clusters

NVIDIA AI; Layer-de-la-Haye; On-site; GBP 120,000 - 180,000; Full time

Job summary

NVIDIA AI in the United Kingdom seeks a Senior Solutions Architect to lead large-scale AI inference initiatives. You will collaborate with EMEA AI Natives, infrastructure providers, and enterprises deploying AI at scale to ensure peak performance and reliability of NVIDIA's inference stack.

You will drive multi-node MoE serving, interconnect-aware scheduling, memory-bound workload optimization, and next-gen architectures, while guiding customers and ML engineers at deep technical level.

Qualifications

• MS or PhD in Computer Science, Engineering, HPC or equivalent.
• 5+ years of Neural Networks inference optimization experience.
• Strong understanding of transformers inference optimization techniques.

Responsibilities

• Guide EMEA AI Natives customers in deploying and optimizing large-scale inference on mult-node GPU clusters.
• Architect efficient inference pipelines for dense and sparse MoE models across thousands of GPUs.
• Improve inference efficiency with quantization, speculative decoding, and KV cache optimization.
• Collaborate with NVIDIA product teams to accelerate customer success.
• Lead workshops and reference architectures across EMEA for the inference stack.

Skills

NN inference optimization

Transformers inference optimization

MoE inference at scale

Expert parallelism

KV cache optimization

Education

MS or PhD in Computer Science, Engineering, HPC or equivalent

Job description

NVIDIA AI in the United Kingdom seeks a Senior Solutions Architect to lead large-scale AI inference initiatives. You will collaborate with EMEA AI Natives, infrastructure providers, and enterprises deploying AI at scale to ensure peak performance and reliability of NVIDIA's inference stack.

You will drive multi-node MoE serving, interconnect-aware scheduling, memory-bound workload optimization, and next-gen architectures, while guiding customers and ML engineers at deep technical level.