Senior Software Engineer — cuEquivariance

Nvidia

Actively hiring
Santa Clara, CA Posted 10 days ago $184,000$287,500 / year

At a glance

AI generated

TL;DR

As a Senior Software Engineer joining NVIDIA BioNeMo’s cuEquivariance team, you will develop and optimize CUDA kernels for advanced biological and scientific applications, focusing on tensor products, segmented polynomials, and triangle-based operations. Your daily tasks include implementing high-performance geometric ML primitives, maintaining interfaces for PyTorch and JAX, and driving CI/CD infrastructure to ensure seamless integration into production pipelines. You will collaborate with research teams and external developers to evaluate new architectures and translate prototypes into robust kernels. Ideal candidates have over six years of experience in CUDA and GPU programming, proficiency in C++ and Python, and a solid understanding of geometric machine learning concepts such as equivariance and group representations. Experience with major neural network frameworks like e3nn or MACE is highly valued, along with contributions to open-source projects in the field.

Skills

CUDA C++ Python PyTorch JAX CI/CD TensorFlow Git GitHub Linux NVIDIA_GPUs Geometric_Machine_Learning Equivariance Group_Representations Memory_Hierarchy Warp_Level_Execution Occupancy Performance_Profiling Mixed_Precision Tensor_Core_Aware_Design

What you'll do

  • Build and optimize CUDA kernels for equivariant neural network primitives.
  • Ensure end-to-end delivery of production-quality software for external frameworks.
  • Develop and maintain PyTorch and JAX interfaces for cuEquivariance primitives.
  • Drive CI/CD infrastructure for multi-GPU kernel builds and performance testing.
  • Collaborate with research teams to translate new architectures into production kernels.
  • Engage with third-party developers to ensure seamless integration of delivered software.

What we're looking for

  • 6+ years of software engineering experience with CUDA and GPU programming.
  • Proficient in C++ and Python for building production libraries used by external developers.
  • Strong foundation in GPU computing, including memory hierarchy and performance profiling.
  • Experience with major neural network frameworks respecting equivariance (e3nn, MACE, etc.).
  • Familiarity with geometric machine learning concepts like equivariance and tensor products.
  • BS/MS in Computer Science, Physics, or related field; PhD preferred in relevant areas.

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 825 open roles on FindRole.

Listed pay typically runs $184,000–$287,500 across 813 roles with salary data.

Most-posted roles

View all roles at Nvidia