System Software Engineer (GPU)

Location: Redwood City, California, United States; on-site

Company

Develops personal home robots designed to automate repetitive household tasks and make generalized robotics broadly accessible.

What you will do

• Own and contribute to the GPU and accelerated compute layer supporting robot perception, ML, controls, and behavior.
• Reduce GPU kernel launch overhead and make model switching fast and predictable.
• Design GPU scheduling and time-slicing for concurrent inference, SLAM, and robotics workloads.
• Build low-latency camera pipelines and CPU–GPU data transfer using pinned memory, zero-copy paths, and asynchronous transfers.
• Design CPU/GPU synchronization patterns that minimize stalls and maintain full inference pipelines.
• Partner with ML, SLAM/Perception, Controls, Hardware, runtime, and build infrastructure teams.

Requirements

• 2+ years of experience developing GPU systems software.
• Strong proficiency in CUDA and a systems language such as C++, C, or Rust.
• Understanding of GPU architecture, GPU workloads, device sharing, and time-slicing trade-offs.
• Hands-on experience with the CUDA Runtime API, CUDA Graphs, and CUDA IPC.
• Familiarity with MPS, MIG, Nsight Systems, and Nsight Compute.
• Solid Linux fundamentals, including scheduling, IPC, memory management, and performance tuning.

Nice to have

• Contributions to CUDA or other GPU programming libraries.
• Camera pipeline integration and NVDEC/NVENC experience.
• Model inference optimization on embedded GPU platforms such as Jetson.
• Observability and tracing for GPU-accelerated workloads.

Culture & Benefits

• Work on personal robotics intended for real household use.
• Collaborate with a curious, creative, and diverse group.
• Equal opportunity employment.