below marketmost similar roles pay hereabove market
This role pays less than
91%
of similar roles. Most pay
$187,850–$235,750
— the blue band above.
At the midpoint, this role pays about
$160k
versus about
$212k
for comparable roles.
Based on 240 similar postings.
Employer
About Nvidia
Nvidia is a leading designer of graphics processing units (GPUs) and system-on-chip units, powering gaming, professional visualization, data centers, and artificial intelligence workloads. Industry: Semiconductors & AI Computing
Nvidia currently has
1623 open roles
on FindRole.
Listed pay typically runs
$184,000–$287,500
across 1110 roles with salary data.
The System Software Engineer, Performance - CUDA Driver joins the team responsible for the CUDA driver and runtime. This role involves designing, implementing, and shipping production C/C++ features and optimizations to improve latency, throughput, and scalability. You will optimize critical execution paths such as kernel launch, synchronization, memory management, and CPU-GPU coordination. Key responsibilities include tracing workloads across application, operating-system, CPU, interconnect, and GPU boundaries to isolate root causes and implement production solutions. Required skills include strong production C/C++ systems-programming, operating systems foundations like threads and virtual memory, and computer-architecture knowledge of memory hierarchy and caching. The work focuses on the CUDA layer to reduce overhead in fundamental data-movement paths, directly impacting performance for AI, deep learning, HPC, graphics, and robotics products.
What does a System Software Engineer earn in California?
Median $235750 from 70 postings across 4 companies.
Design, implement, and ship performance-centric features and programming-model capabilities in the CUDA driver and runtime using C/C++.
Optimize critical execution paths for kernel launch, synchronization, memory management, and CPU-GPU coordination.
Own complex performance problems by forming hypotheses, creating measurements, and isolating root causes across software and hardware boundaries.
Characterize new silicon and establish performance expectations for current and future platforms.
Translate workload evidence into CUDA API improvements and measurement-backed recommendations for future hardware architecture.
Lead cross-layer investigations and define technical direction for major subsystems and hardware/software co-design.
Mentor engineers and conduct design and code reviews to ensure high engineering quality.
What we're looking for
Hold a BS, MS, or PhD in Computer Science, Computer Engineering, Electrical Engineering, or a related field, or equivalent practical experience.
Have at least 2 years of relevant systems-software development experience.
Possess strong production C/C++ systems-programming experience, including delivering substantial features or optimizations in complex codebases.
Demonstrate strong operating systems and concurrency foundations, including threads, synchronization, processes, virtual memory, and user/kernel interactions.
Demonstrate strong computer-architecture foundations, including processors, memory hierarchy, caching, coherence, data movement, and system interconnects.
Show success improving real software performance by measuring behavior, identifying limiting mechanisms, and implementing validated solutions.
Experience developing GPU or accelerator drivers, runtimes, kernel software, firmware, compilers, or other performance-critical low-level systems (preferred).
Experience with pre-silicon analysis, platform bring-up, performance modeling, or hardware/software co-design (preferred).
CUDA
C++
C
GPU
systems-programming
Operating Systems
Computer Architecture
Concurrency
Kernel Software
Drivers
performance-engineering
Python
HPC
Deep Learning
Memory Management
Profiling
Hardware/Software Co-design
CUDA
C
C++
Kernel Mode Drivers
Operating Systems
GPU Architecture
HW/SW Co-design
Virtual Memory
Memory Hierarchy
Interrupts
Memory-Mapped IO
Systems Software Development
Software Verification
High-Performance Computing
CUDA
C
C++
Kernel Mode Drivers
Operating Systems
HW/SW Co-design
Virtual Memory
Threads
Process Control
GPU Architecture
Memory Coherence
Memory-Mapped IO
Software Verification
High-Performance Computing
Artificial Intelligence
C
C++
CUDA
Parallel Computing
Systems Software
Linux
Windows
Kernel Mode Development
Virtual Memory
Memory Hierarchy
Interconnects
Memory-Mapped IO
Operating System Interfaces
CUDA
C
C++
Kernel Programming
Operating Systems
Parallel Programming
Driver Development
Linux
Windows
macOS
System Architecture
Emulation
Virtual Memory
SoC
Hardware Architecture