Job description

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 build the resilience and failure-attribution stack: detecting, triaging, and attributing node, fabric, and workload failures across the cluster at scale. • Build repeatable benchmark suites, automation, acceptance criteria, and qualification workflows on new platforms. • Tune runtime settings, communication parameters, and deployment configurations in close partnership with framework, systems, and platform teams. • Deliver actionable, data-driven recommendations based on profiling, benchmark results, and cluster characterization. • Mentor engineers, drive technical standards, and act as a force multiplier across the broader performance and infrastructure organization. What we need to see: • Bachelor’s or Master’s in Computer Science or a related technical field (or equivalent experience). • 8+ years of experience developing software infrastructure for large-scale AI or HPC systems, including a track record of technical leadership. • Expertise debugging and triaging AI applications across the full stack — from the application layer down to the hardware. • Deep hands-on experience with NCCL, CUDA-aware distributed execution, and debugging multi-GPU and multi-node workloads at scale. • Proven track record of architecting, debugging, and scaling large-scale distributed systems. • Expert-level Python and C/C++ programming skills. • Experience operating workloads in scheduled, containerized cluster environments. • Excellent analytical, debugging, and communication skills, with the ability to influence across teams. Ways to stand out from the crowd: • Demonstrated experience debugging and optimizing AI workloads at large scale. • Deep familiarity with the RDMA software stack (NCCL, IB verbs, UCX, libfabric). • Strong knowledge of GPU cluster fabrics and topology, including NVLink, NVSwitch, PCIe, RoCE, and InfiniBand. • Experience building acceptance tests, benchmark harnesses, regression gates, or cluster qualification tooling for AI platforms. • Experience building resilience, fault-detection, or failure-attribution systems for datacenter-scale infrastructure. NVIDIA is widely considered to be one of the technology world’s most desirable employers. We have some of the most forward-thinking and hardworking people in the world working for us. If you’re creative, autonomous, and love a challenge, we want to hear from you. Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 184,000 USD - 287,500 USD for Level 4, and 224,000 USD - 356,500 USD for Level 5. You will also be eligible for equity and benefits . Applications for this job will be accepted at least until October 3, 2026. This posting is for an existing vacancy. NVIDIA uses AI tools in its recruiting processes. NVIDIA is committed to fostering an inclusive work environment and proud to be an equal opportunity employer. As we highly value diversity in our current and future employees, we do not discriminate (including in our hiring and promotion practices) on the basis of race, religion, color, national origin, gender, gender expression, sexual orientation, age, marital status, veteran status, disability status or any other characteristic protected by law.

Requirements

Must-haves

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 build the resilience and failure-attribution stack: detecting, triaging, and attributing node, fabric, and workload failures across the cluster at scale.
Build repeatable benchmark suites, automation, acceptance criteria, and qualification workflows on new platforms.
Tune runtime settings, communication parameters, and deployment configurations in close partnership with framework, systems, and platform teams.
Deliver actionable, data-driven recommendations based on profiling, benchmark results, and cluster characterization.
Mentor engineers, drive technical standards, and act as a force multiplier across the broader performance and infrastructure organization.
Bachelor’s or Master’s in Computer Science or a related technical field (or equivalent experience).
8+ years of experience developing software infrastructure for large-scale AI or HPC systems, including a track record of technical leadership.
Expertise debugging and triaging AI applications across the full stack — from the application layer down to the hardware.
Deep hands-on experience with NCCL, CUDA-aware distributed execution, and debugging multi-GPU and multi-node workloads at scale.
Proven track record of architecting, debugging, and scaling large-scale distributed systems.
Expert-level Python and C/C++ programming skills.
Experience operating workloads in scheduled, containerized cluster environments.
Excellent analytical, debugging, and communication skills, with the ability to influence across teams.
Demonstrated experience debugging and optimizing AI workloads at large scale.
Deep familiarity with the RDMA software stack (NCCL, IB verbs, UCX, libfabric).
Strong knowledge of GPU cluster fabrics and topology, including NVLink, NVSwitch, PCIe, RoCE, and InfiniBand.
Experience building acceptance tests, benchmark harnesses, regression gates, or cluster qualification tooling for AI platforms.
Experience building resilience, fault-detection, or failure-attribution systems for datacenter-scale infrastructure.
You will also be eligible for equity and benefits .

Nice-to-haves

None flagged.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aillmpython

Platform: eightfold
Effort: medium