GPU Performance Engineer - Neural Reconstruction
Core
Optimizing GPU-accelerated neural reconstruction workflows (Gaussian Splatting, NeRF) for 3D world representation and rendering.
Role type
Senior IC GPU Performance Engineer (Neural Rendering)
Builds
Production-quality neural reconstruction systems for robotics, simulation, and embodied AI.
Domain
Computer Graphics / 3D Vision / GPU Computing
Deliverable
production ML models
Required skills
CUDA programming, C++, Python, PyTorch, GPU profiling, memory optimization, kernel launch optimization, sparse rendering, multiview batching
Preferred skills
Gaussian Splatting, NeRF, differentiable rendering, SLAM, camera/lidar geometry, custom PyTorch operators, distributed training
Technologies
CUDA, PyTorch, C++, Nsight Systems, Nsight Compute, NVTX, NumPy
Responsibilities
Profile end-to-end neural reconstruction workflows to identify bottlenecks; Optimize CUDA/PyTorch performance for rendering and training; Analyze GPU performance using profiling tools; Translate Python bottlenecks into efficient C++/CUDA; Validate performance improvements preserve reconstruction quality; Build repeatable benchmarks and regression tests.
Seniority
Senior, hands-on IC
