Software Engineer, DL Kernel Libraries

About the Role

We're looking for Software Engineer Interns to help develop groundbreaking technologies in the deep learning inference and training systems software stack. The DL Kernel Library team builds innovative AI systems software, primarily cuDNN and FlashInfer.

What You'll Be Doing

• Innovating and developing new AI systems technologies for efficient inference and training
• Designing, implementing, and optimizing kernels for high impact AI workloads
• Designing and implementing extensible abstractions for serving engines and training frameworks
• Building efficient just-in-time domain specific compilers and runtimes
• Collaborating closely with other engineers at NVIDIA across deep learning frameworks, libraries, kernels, and GPU arch teams
• Contributing to open source communities across FlashInfer, cuDNN, vLLM, SGLang, PyTorch, JAX, and XLA

Requirements & Tech Stack

• BS, MS, or PhD degree in Computer Science, Electrical Engineering, or related field (or equivalent experience)
• Strong experience in developing or using deep learning frameworks (e.g. PyTorch, JAX, TensorFlow, ONNX, etc) or inference engines and runtimes such as vLLM, SGLang, and MLC.
• Strong Python and C/C++ programming skills

Ways To Stand Out

• Background in domain specific compiler and library solutions for LLM inference and training (e.g. FlashInfer, Flash Attention)
• Expertise in training frameworks and inference engines
• Expertise in machine learning compilers (e.g. Apache TVM, MLIR)
• Strong experience in GPU kernel development and performance optimizations (especially using CUDA C/C++, CUTLASS DSL, cuTile, Triton, or similar)
• Open source project ownership or contributions