AI Computing Software Intern, GPU Kernel Libraries - 2027
Salary not listed • China, Shanghai • Internship
Posted 1mo ago
Remote Work Policy
On-site
Employment Type
Internship
Categories
AI Infrastructure Engineer
About the job
NVIDIA is seeking passionate Compute/DL Architecture Performance Optimization Interns to explore computer architectures for deep learning and work at the intersection of hardware and software. The role involves mining the ultimate performance of deep learning operators and designing scalable infrastructure to ship highly optimized operators to NVIDIA's software libraries for training and inference.
Responsibilities
- Develop high-performance operators on NVIDIA GPUs for libraries like cuBLAS, TensorRT, cuDNN, cuSparse, and cuTensor.
- Analyze GPU kernel performance on existing and new architectures, identify bottlenecks, and propose solutions.
- Design and develop software for kernel authoring and shipping.
- Apply cutting-edge AI technologies in GPU kernel development workflows.
Requirements
- Currently pursuing a B.S., M.S., or PhD degree in computer science or a related field.
- Strong programming skills in C/C++ and Python.
- Familiarity with GPU programming models and CUDA.
- Good understanding of compiler technologies and experience with LLVM and MLIR.
- Excellent problem-solving, communication, and teamwork skills.