Physical Design Methodology Engineer, Innovus Flows, New College Graduate
NVIDIA
AssemblyPerlPython
Senior Performance Software Engineer, Deep Learning Libraries
NVIDIA
full-time
Posted on:
Origin: • 🇨🇳 China
Visit company websiteJob Level
Senior
Tech Stack
AssemblyTensorflow
About the role
- Develop optimized code to accelerate linear algebra and deep learning operations on NVIDIA GPUs (cuDNN, cuBLAS, TensorRT)
- Write highly tuned compute kernels (matrix multiplies, convolutions, normalizations)
- Tune parallel algorithms and analyze performance to achieve peak GPU efficiency
- Follow software engineering best practices including regression testing and CI/CD flows
- Collaborate with CUDA compiler team, deep learning performance teams, and hardware/architecture teams
- Work on low-level code close to GPU hardware and programming model for new features
Requirements
- Masters or PhD degree or equivalent experience in Computer Science, Computer Engineering, Applied Math, or related field
- 2+ years of relevant industry experience
- Demonstrated strong C++ programming and software design skills, including debugging, performance analysis, and test design
- Experience with performance-oriented parallel programming (e.g., OpenMP or pthreads)
- Solid understanding of computer architecture and some experience with assembly programming
- Ability to identify bottlenecks, optimize resource utilization, and improve throughput
- (Nice to have) Tuning BLAS or deep learning library kernel code
- (Nice to have) CUDA GPU programming
- (Nice to have) Numerical methods and linear algebra
- (Nice to have) LLVM, TVM tensor expressions, or TensorFlow MLIR
