Lead Distributed Deep Learning Communications Architect

NVIDIA; Austin (TX); On-site; USD 184,000 - 287,500; Full time

Benefits offered by this job

Generous benefits package

Job summary

A leading technology company in Austin, Texas, seeks a Deep Learning Communication Architect to optimize data transfer and synchronization in distributed deep learning. This role requires deep knowledge of DNN frameworks, programming expertise in C++ and Python, and at least 6 years of experience in deep learning. The company offers competitive salaries ranging from $184,000 to $356,500, along with generous benefits. Applications are accepted until March 16, 2026.

Qualifications

• 6+ years of experience building and scaling deep neural networks (DNNs).
• Experience optimizing LLM training and inference performance.
• Strong understanding of parallelism techniques such as Data, Pipeline, and Tensor Parallelism.

Responsibilities

• Optimize communication performance in distributed deep learning training.
• Design and implement efficient communication protocols.
• Collaborate with hardware and software teams to enhance system performance.

Skills

Deep learning frameworks (PyTorch, TensorRT‑LLM, vLLM, SGLang)

Programming skills in C++

Programming skills in Python

GPU computing (CUDA, OpenCL)

Education

Ph.D., Master’s, or BS in Computer Science, Electrical Engineering, or related field

Tools

NVLink

InfiniBand

MPI

NCCL

UCX

UCC

NVSHMEM

Job description

A leading technology company in Austin, Texas, seeks a Deep Learning Communication Architect to optimize data transfer and synchronization in distributed deep learning. This role requires deep knowledge of DNN frameworks, programming expertise in C++ and Python, and at least 6 years of experience in deep learning. The company offers competitive salaries ranging from $184,000 to $356,500, along with generous benefits. Applications are accepted until March 16, 2026.