This role pays more than
98%
of similar roles. Most pay
$168,912–$225,400
— the shaded band above.
At the midpoint, this role pays about
$290k
versus about
$197k
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
942 open roles
on FindRole.
Listed pay typically runs
$184,000–$287,500
across 931 roles with salary data.
As a Senior Software Engineer on our Automotive Platform Team, you will lead the technical strategy and architecture for optimizing inference workloads in autonomous driving applications, focusing on deploying deep neural networks efficiently on NVIDIA automotive compute platforms. Your day-to-day responsibilities include driving performance analysis across DNN models, TensorRT/compiler flows, CUDA kernels, memory behavior, scheduling, runtime services, and platform constraints while developing model optimization techniques such as quantization, pruning, distillation, graph optimization, operator fusion, kernel selection, and layout/memory optimization. You will collaborate closely with teams specializing in TensorRT, CUDA, compiler technology, silicon architecture, perception, planning, DriveOS, and safety platforms to build tools and methodologies for profiling, benchmarking, debugging, and validating model and platform performance. The ideal candidate has extensive experience in systems software, AI/ML infrastructure, deep learning inference, and compiler/runtime technology, with strong skills in C/C++ and Python, along with familiarity with TensorRT, ONNX, PyTorch, CUDA, Triton, and related frameworks. Experience deploying optimized DNNs on embedded or automotive platforms and a background in autonomous driving, ADAS, robotics, real-time systems, safety-aware software, or deterministic low-latency systems is highly desirable.
Lead architecture and technical strategy for optimizing inference workloads in autonomous driving applications.
Develop model optimization techniques including quantization, pruning, distillation, graph optimization, operator fusion, kernel selection, and layout/memory optimization.
Conduct end-to-end performance analysis across DNN models, TensorRT/compiler flows, CUDA kernels, memory behavior, scheduling, runtime services, and automotive platform constraints.
Build profiling, benchmarking, debugging, and validation tools for model and platform performance.
Collaborate with TensorRT, CUDA, compiler, silicon architecture, perception, planning, DriveOS, and safety platform teams to integrate optimization techniques.
What we're looking for
Over 12 years of experience in systems software, AI/ML infrastructure, and deep learning inference.
Expertise in C/C++ and practical Python for software development.
Deep knowledge of TensorRT, CUDA, Triton, ONNX, PyTorch, and related frameworks.
Hands-on experience optimizing DNN models for latency, throughput, memory footprint, and power.
Familiarity with deploying optimized deep learning models on automotive platforms.
C++
CUDA
Python
ISO 26262
ASIL
TensorRT
Modern C++ standards
deep learning frameworks
performance benchmarking
profiling
optimizations
systems programming
embedded systems
compiler development