CUDA Driver & Runtime Systems Engineer

Socket.dev; Santa Clara (CA); On-site; USD 124,000 - 196,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking systems software engineers in Santa Clara, CA to design and ship production C/C++ features for the CUDA driver and runtime. You will optimize critical execution paths, memory management, and interconnect use while collaborating across teams to drive performance improvements for AI, HPC, and graphics workloads.

The role requires strong systems programming, OS concurrency, and computer-architecture knowledge, with deep problem solving and clear communication.

Qualifications

• BS/MS/PhD in Computer Science, Computer Engineering, Electrical Engineering, or related field—or equivalent practical experience.
• At least 2 years of systems-software development experience.
• Strong production C/C++ systems-programming experience in a complex codebase.
• Solid OS and concurrency foundations: threads, synchronization, processes, VM, user/kernel interactions.
• Strong computer-architecture foundations: processors, memory hierarchy, data movement, interconnects.
• Proven ability to measure, analyze, and improve software performance.
• Sound technical judgment and clear communication across teams.

Responsibilities

• Design, implement, validate, and ship performance-centric C/C++ features in the CUDA driver and runtime.
• Optimize kernel launch, synchronization, memory movement, CPU-GPU coordination, and interconnect usage for latency and throughput.
• Own complex performance problems end-to-end, form hypotheses, measure, and validate impact.
• Set performance expectations for current and future platforms; drive silicon/software alignment.
• Translate workload evidence into CUDA API improvements and hardware direction.
• Collaborate with application, library, OS, driver, and customer-facing teams; participate in design reviews.

Skills

C/C++ systems programming

Concurrency & OS fundamentals

Computer architecture foundations

Performance engineering

Communication & ownership

CUDA/GPU experience (nice to have)

Education

BS/MS/PhD in CS/CE/EE or equivalent

Tools

CUDA driver/runtime

Python scripting

Job description

NVIDIA is seeking systems software engineers in Santa Clara, CA to design and ship production C/C++ features for the CUDA driver and runtime. You will optimize critical execution paths, memory management, and interconnect use while collaborating across teams to drive performance improvements for AI, HPC, and graphics workloads.

The role requires strong systems programming, OS concurrency, and computer-architecture knowledge, with deep problem solving and clear communication.