Nvidia

Software Engineer, Collective Communication and Network Performance

Nvidia
Apply
2 days ago
Austin, TX, USA or Santa Clara, CA, USASenior
H1B sponsor

Base Salary

$173k - $338k/yr

Responsibilities

  • Develop detection algorithms that identify collective communication performance problems from always-on telemetry.
  • Build infrastructure that ingests per-collective records, correlates network counters and topology, and produces diagnoses with supporting evidence.
  • Define required data from NCCL profiler and tuner plugin interfaces, RDMA transport state, and NIC and switch counters.
  • Reproduce communication problems on internal clusters and maintain regression tests that validate automated diagnoses.
  • Extend diagnostic coverage beyond NCCL to point-to-point and other communication paths.

Requirements

  • Bachelor's degree in Computer Science, Computer Engineering, or a related field, or equivalent experience.
  • 8+ years of relevant software engineering experience, including hands-on work with a collective communication library, RDMA transport, or GPU/HPC network performance.
  • Working knowledge of NCCL or a comparable library, including algorithms, protocols, channels, transports, topology detection, latency, and bandwidth.
  • Practical knowledge of RDMA networking, including InfiniBand or RoCE, NIC and switch counters, congestion control, and measurement tools.
  • Experience diagnosing multi-node communication problems from measurements and converting diagnoses into software used by others.
  • History of delivering changes in code owned by other teams, from requirements definition through delivery.
  • Strong cross-component debugging skills and validation through code review, testing, and independent measurements.
  • Proficiency in C++ and Python and ability to communicate evidence, uncertainty, and next steps clearly.
  • Preferred experience contributing to NCCL, UCX, libfabric, NVSHMEM, profiling or tuner plugins, NIXL, DeepEP, or EFA-related systems.
  • Preferred experience with production detection or root-cause analysis, or with operating and debugging training or inference jobs above a thousand GPUs.

Benefits

  • The position is based in Santa Clara, California.
  • Eligible for equity and benefits.

Tech Stack

Categories

Nvidia

About Nvidia

10,000+ employees

Nvidia designs and sells GPUs and accelerated computing platforms for data centers, AI/ML, graphics, gaming, and automotive, monetizing through hardware, software platforms (CUDA, AI frameworks), and systems like DGX and networking. Customers include cloud providers, enterprises, researchers, and OEMs. Founded in 1993 and headquartered in Santa Clara, it is a public company traded on NASDAQ under NVDA.

Contact me