NVIDIA seeks a Senior Deep Learning Performance Architect to innovate deep learning architectures that enhance performance and efficiency. This role involves analyzing hardware-software interactions for future algorithms and applications, developing analytical models and simulators, and collaborating with cross-functional teams to guide the direction of deep learning hardware and software. Ideal candidates hold an MS or PhD in Computer Science or related fields, with over six years of experience in GPU or Deep Learning ASIC architecture. Strong programming skills in Python, C, and C++ are essential, along with expertise in frameworks like PyTorch and libraries such as CUDA and MLIR. The position requires a deep understanding of machine learning, performance modeling, and the ability to think critically about complex architectural challenges in the rapidly evolving AI landscape.
Skills
Python
C
C++
Pytorch
JAX
TensorRT
CUDNN
CUBLAS
CUTLASS
MLIR
Triton
CUDA
OpenCL
GPU
Deep Learning ASIC
Performance Modeling
Architecture Simulation
Profiling
Analysis
What you'll do
Develop innovative deep learning architectures to enhance performance and efficiency.
Analyze hardware and software trade-offs using models, simulators, and test suites.
Guide the direction of deep learning hardware and software through collaboration.
Harness groundbreaking frameworks, libraries, and compilers for deep learning.
Understand future algorithms and programming models' impact on architecture.
What we're looking for
MS or PhD in Computer Science, Engineering with 6+ years of relevant experience
Strong background in GPU or Deep Learning ASIC architecture for training and inference
Experience with performance modeling, simulation, profiling, and analysis
Solid foundation in machine learning and deep learning frameworks (PyTorch, JAX)
Proficient programming skills in Python, C, C++ and knowledge of CUDA, OpenCL
Familiarity with libraries like CUDNN, CUBLAS, CUTLASS, MLIR, Triton for optimization
Market check
Salary context
Above market
How this pay compares to similar roles
Similar $218k
This role $236k
$171k
most similar roles pay here
$300k
This role pays more than
72%
of similar roles. Most pay
$190,218–$246,150
— the shaded band above.
At the midpoint, this role pays about
$236k
versus about
$218k
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
855 open roles
on FindRole.
Listed pay typically runs
$184,000–$287,500
across 843 roles with salary data.
Python
C
C++
CUDA
MPI
OpenMP
HPC
GPU
Deep Learning
Machine Learning
Performance Modeling
Systems Performance Analysis
AI Inference Workloads
CUDA Kernels
Custom ASIC Hardware