NVIDIA
NVIDIA
NVIDIA develops accelerated computing and artificial intelligence technologies used across gaming, data centers, cloud computing, healthcare, manufacturing, automotive, and robotics. Its work spans graphics processing units, AI platforms, simulation, and industry-focused solutions, including NVIDIA Omniverse for collaborative 3D workflows, NVIDIA DRIVE for autonomous vehicle development, and NVIDIA Clara for healthcare applications. The company brings together large technical teams working on the infrastructure and software that support advanced computing, simulation, and AI-driven analytics.

Senior Deep Learning Compiler Engineer – CUDA Tile and LPU

Develop CUDA Tile transformations and MLIR lowering passes for NVIDIA’s AI and GPU computing platforms. Improve tile-based kernel performance across GPU and LPU architectures.

Description

  • Support CUDA Tile, NVIDIA’s tile-based programming model, for LPU development
  • Design and implement compiler transformations
  • Create MLIR dialects and lowering passes
  • Improve tile-based kernel performance across multiple generations of NVIDIA GPU and LPU architectures
  • Define public APIs
  • Develop compiler and optimization methods
  • Handle performance tuning and broader software engineering tasks
  • Collaborate with an expert team on work serving the deep learning community

Requirements

  • Bachelor’s, master’s, or doctoral degree in computer science, computer engineering, or a related discipline, or equivalent experience
  • At least three years of relevant experience or research in compiler optimization, performance analysis, and intermediate representation design
  • Ability to work autonomously, establish project goals and scope, and lead individual development efforts
  • Advanced C and C++ programming and software design skills, including debugging, performance analysis, and test planning
  • Strong communication skills and the ability to contribute effectively within a fast-moving, product-focused team
  • Familiarity with CPU and/or GPU architecture is preferred
  • Experience programming with CUDA or OpenCL is preferred
  • Experience with MLIR, LLVM, XLA, TVM, deep learning models, and deep learning algorithms is preferred

Benefits

  • Equity compensation
  • Comprehensive benefits package

Related Jobs