Senior System Software Engineer – Neural Graphics Performance

About This Role

NVIDIA's Omniverse Neural Reconstruction (NuRec) SDK team is looking for a performance-focused systems engineer to optimize Gaussian Splatting and neural reconstruction pipelines. You will ensure these cutting-edge graphics techniques run at real-time frame rates on NVIDIA GPUs.

What You Will Be Doing

• Profile and optimize GPU-accelerated neural graphics pipelines using CUDA, Slang, and shader languages (GLSL, HLSL, Metal)
• Develop performance benchmarking frameworks and regression testing infrastructure for NuRec SDK components
• Identify and resolve performance bottlenecks in Gaussian Splatting rendering, neural radiance field training, and 3D reconstruction workflows
• Collaborate with research and product teams to balance visual quality with latency and throughput targets
• Design efficient memory management and streaming strategies for large-scale neural scene representations

What We Need to See

• MS in Computer Science, Electrical Engineering, or a related field (or equivalent experience)
• 3+ years of GPU performance engineering experience
• Deep proficiency with CUDA, GPU profiling tools (Nsight, CUPTI), and at least one shading language
• Strong C++ and Python skills with experience in performance-critical codebases
• Familiarity with neural rendering or real-time 3D graphics pipelines

The base salary range is $152,000 – $287,500. Your total rewards package includes equity and comprehensive benefits.