Last year NVIDIA [1]announced the new CUDA Tile programming model as one of the biggest updates ever to the CUDA platform. CUDA Tile brings a virtual ISA for tile-based parallel programming and they subsequently [2]open-sourced the CUDA Tile IR as an intermediate representation built atop LLVM's MLIR. Now they are looking to hire additional LLVM compiler engineers to help foster their CUDA Tile initiatives.

Announced today on the LLVM Discourse is word they are hiring more machine learning compiler engineers for working on CUDA Tile "at all levels". Their technical description characterizes CUDA Tile as being built natively on LLVM's wonderful [3]MLIR with a mix of open-source and proprietary dialects/passes:

"CUDA Tile is built natively on MLIR, leveraging a mix of open-source and NVIDIA-proprietary dialects and passes. Our team includes many prominent contributors and recognized leaders within the MLIR community, offering a world-class environment for compiler innovation."

Experienced compiler engineers familiar with MLIR can learn more about NVIDIA's latest job openings via [4]the LLVM Discourse .



[1] https://www.phoronix.com/news/NVIDIA-CUDA-13.1

[2] https://www.phoronix.com/news/NVIDIA-CUDA-Tile-IR-Open-Source

[3] https://www.phoronix.com/search/MLIR

[4] https://discourse.llvm.org/t/nvidia-is-hiring-compiler-engineers-for-cuda-tile/90519