Guray Ozen

Compiler Research at NVIDIA

gry3.jpg

Guray Ozen is a compiler engineer in the Machine Learning Compiler team at NVIDIA.

Current Projects

Guray currently works on two programming models based on MLIR compiler:

  • TileIR Compiler: A next-generation MLIR-based IR for portable GPU programming with a focus on kernel composition and performance tuning across ML and HPC workloads.
  • CuTeDSL, Cutlass Python DSL: Cutlass 4.0 introduced CuTe Python DSL. We designed this for writing high-performance GPU kernels, integrated with CUTLASS and designed for composable kernel development.

Research Interest

His current focus centers on optimizing compilers and programming languages for GPU utilization in machine learning (ML) and high-performance computing (HPC). He has made key contributions to several production-grade compilers, including Clang, Flang, MLIR, IREE, and NVIDIA HPC (formerly PGI).

Previously, he was actively involved in language design for parallel programming models, such as OpenMP and OpenACC. He served as a voting member of the OpenMP Language Committee for NVIDIA and contributed extensively to the OpenACC language specification.

News

Dec 19, 2025 GitHub TileIR’s cuda-tile MLIR Dialect is now open source
Dec 01, 2025 NVIDIA TileIR compiler is released with Cuda 13.1
Oct 28, 2025 MLIR Talk on CuTeDSL, CUTLASS Python DSL Infrastructure at LLVM’25
Aug 18, 2025 MLIR Now an official maintainer of the NVVM and NVGPU dialects in MLIR! See all maintainers
May 13, 2025 NVIDIA CuTeDSL is released with Cutlass 4.0

Work Experience

  • Compiler Engineer, ML Compilers, NVIDIA, Switzerland (2024 - Present)
  • Compiler Research Engineer, ML Compiler Systems Research, Google Research (DeepMind), Switzerland (2022 - 2024)
  • Compiler Engineer, NVIDIA HPC Compilers, NVIDIA, Germany (2018 - 2022)
  • Compiler Research Intern, NVIDIA, USA (2017)
  • Short-term Researcher, Advanced Compilers Group, IBM T.J. Watson Research Center, Yorktown Heights, USA (2016)
  • Research Assistant, Barcelona Supercomputing Center, Spain (2013 - 2017)
  • Software Engineer, Veripark, Akbank, Istanbul (2010 - 2012)

Education

  • PhD in Computer Architecture (Excellent Cum Laude), Universitat Politècnica de Catalunya (UPC), Spain (2018)
  • MSc High-Performance Computing, Universitat Politècnica de Catalunya (UPC), Spain (2014)
  • BSc Computer Science Engineering, Dokuz Eylul University, Turkey (2010)

Selected talks and publications

2025

  1. LLVM
    llvm.png
    CuTeDSL, CUTLASS Python DSL Infrastructure
    Guray Ozen
    2025
  2. EuroLLVM
    llvm.png
    Bringing NVIDIA Blackwell support to LLVM and MLIR
    Guray Ozen
    2025

2024

  1. ICML
    icml.png
    NVDSL: Simplifying Tensor Cores with Python-Driven MLIR Metaprogramming
    Guray Ozen
    In ESFOMO Workshop at ICML, 2024
  2. EuroLLVM
    llvm.png
    Zero to Hero: Programming Nvidia Hopper Tensor Core with MLIR’s NVGPU Dialect
    Guray Ozen
    2024