Senior AI Inference Architect: Ultra-Scale MoE on GPU Clusters

NVIDIA; Poland; On-site; PLN 293,000 - 507,000; Full time

Job summary

NVIDIA in Poland is seeking a Senior Solutions Architect with deep experience in large-scale production AI inference. You will lead technical direction for scalable, high-performance inference across multi-node GPU deployments and collaborate with EMEA AI Natives, infrastructure providers, and enterprises.

You will optimize inference pipelines, shaping performance, efficiency, and reliability across quantization, MoE workloads, and advanced architectures while engaging ML engineers, researchers,

Qualifications

• MS or PhD in Computer Science, Engineering, High-Performance Computing, or equivalent professional experience.
• 5+ years of experience in Neural Networks inference optimization.
• Solid understanding of transformers inference optimization: quantization, disaggregated inference, speculative decoding, continuous batching, KV cache optimization.
• Practical experience in MoE inference at scale: expert parallelism, WideEP, all-to-all communication, routing overhead, and load balancing at scale.
• Ability to engage effectively with ML engineers, researchers, and systems architects at a deep technical level.

Responsibilities

• Guide EMEA AI Natives customers in deploying and optimizing large-scale inference workloads on multi-node GPU clusters.
• Architect efficient inference pipelines for dense and sparse/latent MoE models distributing workload among thousands of GPUs.
• Improve inference efficiency across quantization (INT4/FP8), speculative decoding, disaggregated prefill/decode, KV cache management, and WideEP for large MoE deployments.
• Collaborate with NVIDIA product teams (Dynamo, TensorRT-LLM, NIXL) to accelerate customer success.
• Animate the AI inference developer’s community across EMEA through technical workshops, hackathons, and reference architectures.

Skills

NN inference optimization

MoE inference at scale

Transformers inference optimization

Engage with ML engineers/researchers

Education

MS or PhD in Computer Science/Engineering or equivalent

Tools

NVIDIA Dynamo

NIXL

TensorRT-LLM

Job description

NVIDIA in Poland is seeking a Senior Solutions Architect with deep experience in large-scale production AI inference. You will lead technical direction for scalable, high-performance inference across multi-node GPU deployments and collaborate with EMEA AI Natives, infrastructure providers, and enterprises.

You will optimize inference pipelines, shaping performance, efficiency, and reliability across quantization, MoE workloads, and advanced architectures while engaging ML engineers, researchers,