Senior Embedded Software Engineer — Edge Compute & GPU

Socket.dev; Sunnyvale (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Intuitive Surgical in Sunnyvale, CA is seeking an experienced embedded software engineer to design, build, and optimize reusable software frameworks for compute-intensive applications on the dV5 system and future platforms.

The role focuses on edge compute architecture, distributed frameworks, data transfer, graph-based execution, and performance engineering for production embedded systems, covering Linux, device drivers, CUDA kernels, inference engines, and middleware.

Qualifications

• Experience designing embedded software infrastructure for edge compute systems.
• Strong Linux system software development and Linux internals knowledge.
• Experience with GPU programming and CUDA.
• Experience with ML inference frameworks (e.g., TensorRT).
• Proficiency in Python and shell scripting.

Responsibilities

• Design and develop multi-threaded C++ embedded Linux software infrastructure across heterogeneous hardware.
• Evolve distributed compute frameworks and graph-based execution pipelines for ML workloads.
• Profile and optimize CPU/GPU pipelines and multi-node systems for performance, latency, and power.
• Create automated performance tests and CI integrations to maintain baselines.
• Collaborate with hardware, platform, and algorithms teams to define reusable frameworks.

Skills

Embedded Linux

C++

CUDA

GPU optimization

Distributed compute

Python

Shell scripting

System architecture

Performance profiling

Education

BS in Computer Science, Electrical/Computer Engineering

MS optional with 5+ years experience

Tools

Nsight Systems

Nsight Compute

eBPF tooling

Job description

Intuitive Surgical in Sunnyvale, CA is seeking an experienced embedded software engineer to design, build, and optimize reusable software frameworks for compute-intensive applications on the dV5 system and future platforms.

The role focuses on edge compute architecture, distributed frameworks, data transfer, graph-based execution, and performance engineering for production embedded systems, covering Linux, device drivers, CUDA kernels, inference engines, and middleware.