Senior Software Engineer, DGX Cloud AI Infrastructure

About this role

NVIDIA is at the forefront of the generative AI revolution, building the software and systems that power the world’s most advanced large language model workloads. We are looking for a Senior Software Engineer to lead the bring-up, triage, benchmarking, analysis, and optimization of distributed training and inference workloads across NVIDIA GPU platforms at the largest scales we run.

In this role you will set technical direction across communication libraries, model frameworks, and inference/training stacks to ensure state-of-the-art LLM workloads run efficiently and reliably at scale. You will lead deep performance and reliability investigations on multi-GPU and multi-node deployments, define how we benchmark and qualify new platforms, and build the resilience and failure-attribution capabilities that keep large clusters productive. This is a hands-on senior individual-contributor role for an engineer who operates at the intersection of deep learning systems, GPU performance, distributed computing, and large-scale operations — and who raises the bar for the engineers around them.

What you’ll be doing:

• Lead bring-up, validation, and debugging of large-scale AI clusters, infrastructure, and end-to-end workloads, setting the standard for how the team operates.
• Bring up, tune, and benchmark AI pre-training, post-training, and inference workloads using PyTorch, NeMo / Megatron, TensorRT-LLM, and adjacent NVIDIA AI software stacks.
• Profile and optimize end-to-end workload performance across compute, memory, networking, and communication layers using tools such as Nsight Systems, NCCL tests, and custom microbenchmarks.
• Analyze scaling efficiency for distributed LLM workloads using data, tensor, pipeline, and expert parallelism across modern GPU clusters, and translate findings into concrete tuning guidance.
• Own root-cause analysis of complex failures — hangs, performance regressions, topology sensitivity in large distributed environments.
• Define and buil...