Performance Engineer - GPU Communication Libraries

Employment Type: Full-time

Position Overview

NVIDIA is a leader in Artificial Intelligence, High-Performance Computing (HPC), and Visualization. Our invention, the GPU, is central to modern computing, driving advancements from AI to autonomous vehicles. We are seeking a motivated Performance Engineer to influence the roadmap of our GPU communication libraries (NCCL, NVSHMEM, GPUDirect). These libraries are critical for scaling Deep Learning and HPC applications, which increasingly demand massive compute power across tens of thousands of GPUs connected via high-speed interconnects and networking. This role offers an exceptional opportunity to advance the state-of-the-art in GPU communication performance for large-scale deployments.

Responsibilities

• Conduct in-depth performance characterization and analysis on large multi-GPU and multi-node clusters.
• Study the interaction of our libraries with all hardware (GPU, CPU, Networking) and software components in the stack.
• Evaluate proof-of-concepts and conduct trade-off analysis for multiple solutions.
• Triage and root-cause performance issues reported by customers.
• Collect and analyze performance data; build tools and infrastructure for visualization and analysis.
• Collaborate with a dynamic, multi-time zone team.

Requirements

• M.S. (or equivalent experience) or Ph.D. in Computer Science or a related field.
• Relevant performance engineering and HPC experience.
• 3+ years of experience with parallel programming.
• Experience with at least one communication runtime (MPI, NCCL, UCX, NVSHMEM).
• Experience conducting performance benchmarking and triage on large-scale HPC clusters.
• Good understanding of computer system architecture, hardware-software interactions, and operating system principles.
• Ability to implement micro-benchmarks in C/C++ and modify codebases.
• Proficiency in debugging performance issues across the entire hardware/software stack.
• Proficiency in a scripting language, preferably Python.
• Familiarity with containers, cloud provisioning, and scheduling tools (Kubernetes, SLURM, Ansible, Docker).
• Adaptability and passion for learning new areas and tools.
• Flexibility to work and communicate effectively across different teams and time zones.

Ways to Stand Out

• Practical experience with InfiniBand/Ethernet networks (RDMA, topologies, congestion control).
• Experience debugging network issues in large-scale deployments.
• Familiarity with CUDA programming and/or GPUs.
• Experience with Deep Learning Frameworks such as PyTorch, TensorFlow.

Company Information

NVIDIA is at the forefront of breakthroughs in Artificial Intelligence, High-Performance Computing, and Visualization. Our teams consist of driven, innovative professionals dedicated to pushing the boundaries of technology. We offer highly competitive salaries, an extensive benefits package, and a work environment that promotes diversity, inclusion, and flexibility. As an equal opportunity employer, we are committed to fostering a supportive and empowering workplace for all.