Senior DGX Cloud Performance Engineer

Details

NVIDIA DGX™ Cloud is an end-to-end, scalable AI platform for developers, offering scalable capacity built on the latest NVIDIA architecture and co-engineered with the world’s leading cloud service providers (CSPs). We are seeking highly skilled parallel and distributed systems engineers to drive the performance analysis, optimization, and modeling to define the architecture and design of NVIDIA's DGX Cloud clusters. The ideal candidate will have a deep understanding of the methodology to conduct end-to-end performance analysis of critical AI applications running on large-scale parallel and distributed systems. Candidates will work closely with multi-functional teams to define DGX Cloud cluster architecture for different CSPs, optimize workloads running on these systems and develop the methodology that will drive the HW–SW codesign cycle to develop elite AI infrastructure at scale and make them more easily consumable by users (via improved scalability, reliability, cleaner abstractions, etc.).

Responsibilities

• Develop benchmarks and end-to-end customer applications running at scale, instrumented for performance measurements, tracking, and sampling, to measure and optimize performance of meaningful applications and services.
• Construct carefully designed experiments to analyze, study and develop critical insights into performance bottlenecks and dependencies from an end-to-end perspective.
• Develop ideas to improve end-to-end system performance and usability by leading changes in hardware or software (or both).
• Collaborate with external cloud service providers (CSPs) during the full lifecycle of cluster deployment and workload optimization to understand and drive standard methodologies.
• Collaborate with AI researchers, developers, and application service providers to understand difficulties, requirements, project future needs and share best practices.
• Work with a diverse set of LLM workloads and their application areas such as healthcare, climate modeling, pharmaceuticals, financial futures, genomics/drug discovery, among others.
• Develop modeling frameworks and TCO analysis to enable efficient exploration and sweep of the architecture and design space.
• Develop the methodology needed to drive engineering analysis to advise the architecture, design and roadmap of DGX Cloud.

Requirements

• 12+ years of proven experience.
• Ability to work with large-scale parallel and distributed accelerator-based systems.
• Expertise optimizing performance and AI workloads on large-scale systems.
• Experience with performance modeling and benchmarking at scale.
• Strong background in computer architecture, networking, storage systems, and accelerators.
• Familiarity with popular AI frameworks: PyTorch, TensorFlow, JAX, Megatron-LM, Tensort-LLM, VLLM, among others.
• Experience with AI/ML models and workloads, in particular large language models (LLMs).
• Understanding of deep neural networks (DNNs) and their use in emerging AI/ML applications and services.
• Bachelors or Masters in Engineering (preferably Electrical Engineering, Computer Engineering, or Computer Science) or equivalent experience.
• Proficiency in Python and C/C++.
• Expertise with at least one public cloud infrastructure provider (GCP, AWS, Azure, OCI, …).

Ways to Stand Out

• Very high intellectual curiosity; confidence to dig in as needed; not afraid of confronting complexity; able to pick up new areas quickly.
• Proficiency in CUDA and XLA.
• Excellent interpersonal skills.
• PhD is a nice-to-have.

Compensation & Benefits

Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 224,000 USD - 356,500 USD for Level 5, and 272,000 USD - 425,500 USD for Level 6. You will also be eligible for equity and benefits.

Additional

NVIDIA is committed to fostering a diverse work environment and is an equal opportunity employer. For more details about benefits, see: http://www.nvidiabenefits.com/ and https://www.nvidia.com/en-us/benefits/.