Distinguished Software Architect, Deep Learning and HPC Communications

Nvidia

Confirmed live yesterday High trust

Quick summary

Work type
On-site
Location
Santa Clara, CA
Salary
$320,000–$488,750 / yr
Posted
6 days ago
Freshness
Confirmed live yesterday

Market check

Salary context

Above market

How this pay compares to similar roles

Similar $232k
This role $404k
$147k most similar roles pay here $525k

This role pays more than 99% of similar roles. Most pay $196,750–$266,375 — the shaded band above. At the midpoint, this role pays about $404k versus about $232k for comparable roles.

Based on 240 similar postings.

Employer

About Nvidia

Nvidia is a leading designer of graphics processing units (GPUs) and system-on-chip units, powering gaming, professional visualization, data centers, and artificial intelligence workloads. Industry: Semiconductors & AI Computing

Nvidia currently has 896 open roles on FindRole.

Listed pay typically runs $184,000–$287,500 across 876 roles with salary data.

Most-posted roles

View all roles at Nvidia

At a glance

TL;DR · Distinguished Software Architect, Deep Learning and HPC Communications

The Distinguished Software Architect - Deep Learning and HPC Communications joins the GPU Communications Libraries and Networking team to co-design next-generation data center platforms. This role involves researching new communication technologies, expanding the GPUDirect portfolio, and designing features for communication libraries like NCCL, NVSHMEM, and UCX. The architect will propose hardware and software solutions, analyze technical models to drive changes based on quantitative data, and promote new communication technologies across application verticals. Key requirements include expertise in HPC, parallel programming models such as MPI and SHMEM, GPU architecture, CUDA, and high-performance networking including Infiniband and Ethernet. Candidates must possess proficiency in C or C++ for systems software development and deep knowledge of ML/DL fundamentals. The role addresses the critical challenge of optimizing communication performance across tens of thousands of GPUs connected via NVLink and PCIe.

What you'll do

  • Research new communication technologies and expand the GPUDirect technology portfolio.
  • Design new features for GPU communication libraries like NCCL, NVSHMEM, and UCX.
  • Propose innovative hardware and software solutions for next-generation data center platforms.
  • Co-design technical solutions with GPU, Networking, and Software architects to ensure seamless integration.
  • Drive architectural changes based on quantitative data from proof-of-concepts and technical modeling.
  • Drive the adoption of new communication technologies across various application verticals.
  • Monitor and integrate the latest deep learning research into existing software stacks.

What we're looking for

  • PhD in Computer Science, Computer Engineering, or a related field is required, or strong equivalent experience.
  • 15+ years of relevant experience in academia or the industry are required.
  • Expertise in HPC, parallel programming models (MPI, SHMEM), and at least one communication runtime like NCCL or UCX is required.
  • Deep knowledge of computer/system architecture, GPU architecture, and CUDA is required.
  • Extensive experience in high performance networking including Infiniband, Ethernet, network design, and performance analysis is required.
  • Proficiency in C or C++ for systems software development is required.
  • Strong skills in ML/DL fundamentals, parallel algorithms, fault tolerance, and developing applications using PyTorch or TensorFlow are preferred.
  • Candidates should be recognized leaders in HPC/DL communications with a history of patents, publications, or influence in industry standards.

More like this

Similar roles

Senior Deep Learning Communication Architect

Nvidia

Santa Clara, CA +1 114 days ago $184,000$287,500
Deep Learning PyTorch TensorRT-LLM vLLM SGLang C++ Python CUDA OpenCL InfiniBand NVLink NCCL MPI UCX UCC NVSHMEM RoCE Triton FSDP
6+ yrs exp

Principal Deep Learning Communication Architect

Nvidia

Remote (Santa Clara, CA) +1 150 days ago $272,000$431,250
NCCL UCX UCC NVSHMEM CUDA InfiniBand RDMA RoCE TensorRT-LLM vLLM SGLang Megatron-Core DeepSpeed JAX XLA PyTorch Distributed Ray HPC High-Performance Computing
10+ yrs exp Remote