Embedded / GPU Platform Engineer

Ensigncode is hiring an Embedded/GPU Platform Engineer to own deployment and device reliability: integrating GPU software with target systems, diagnosing driver, CUDA runtime, and device-level issues, and building repeatable deployment and health-check workflows. CUDA or NVIDIA platform exposure is a strong advantage; general embedded experience alone is not a substitute.

Optimized GPU code that only runs reliably on a developer’s own workstation isn’t actually done. As Embedded/GPU Platform Engineer, you’ll own deployment and device reliability: integrating our GPU software with target systems, diagnosing driver, CUDA runtime, and device-level issues, and building the logging, health checks, and troubleshooting workflows that make problems diagnosable outside the dev environment, including on edge and embedded deployments where applicable.

We’re specifically looking for CUDA or NVIDIA platform exposure here. General embedded systems experience is valuable, but it complements GPU fundamentals rather than substituting for them.

Responsibilities

• Integrate GPU software with target systems and deployment environments; manage device/runtime dependencies and configuration
• Diagnose platform-level issues involving drivers, CUDA runtime, device availability, memory limits, power/thermal constraints, and system stability
• Build repeatable deployment, logging, health checks, and troubleshooting workflows
• Support edge or embedded deployments where applicable; collaborate on device interfaces and system-level constraints

Core requirements

• C++ and Linux; embedded/system software; device integration; debugging and deployment practices
• CUDA/NVIDIA platform exposure is a strong advantage; embedded experience should complement, not replace, GPU fundamentals

Team contribution

Ensures optimized software runs reliably on the intended hardware and is diagnosable outside the development workstation.

Works most closely with