Senior Rust-to-GPU Compiler Architect

NVIDIA; Washington (Washington County); On-site; USD 152,000 - 241,500; Full time

Job summary

NVIDIA is seeking a Senior Compiler Engineer to advance Rust-to-GPU compilation and native IR frameworks for CUDA and PTX. You will craft pipelines, backends, and optimizers that translate Rust into high-performance GPU code, enabling safer, more expressive GPU kernels.

You will work on advanced compiler architectures, JIT/AOT flows, and GPU-aware tooling, partnering with teams across NVIDIA to push the boundaries of accelerated computing.

Qualifications

• 5+ years of compiler development or GPU code generation experience.
• Deep expertise in Rust and compiler internals (rustc, MIR, borrow-checker).
• Experience with IRs and code generation (LLVM MLIR or custom IR).
• Understanding of CUDA GPU architectures and parallel programming.

Responsibilities

• Design and implement Rust-to-GPU compiler backends and Rust-native IR frameworks.
• Build compiler IRs and optimizers for GPU target pipelines (MLIR/PTX/LLVM).
• Support ahead-of-time, JIT, and link-time optimization workflows for NVIDIA GPUs.
• Define safe parallel abstractions across GPU launch boundaries to ensure memory safety.
• Expose device-side abstractions for low-level GPU primitives in Rust.
• Collaborate across teams to deliver scalable compiler-engineering成果.

Skills

Rust expertise

Compiler internals

IR frameworks

CUDA/GPU programming

LLVM/MLIR

Performance analysis

Python/ scripting

Code optimization

Education

Bachelor's/Master's/Ph.D. in CS/CE or equivalent

Tools

LLVM IR

MLIR

Rust MIR

proc-macros

Job description

NVIDIA is seeking a Senior Compiler Engineer to advance Rust-to-GPU compilation and native IR frameworks for CUDA and PTX. You will craft pipelines, backends, and optimizers that translate Rust into high-performance GPU code, enabling safer, more expressive GPU kernels.

You will work on advanced compiler architectures, JIT/AOT flows, and GPU-aware tooling, partnering with teams across NVIDIA to push the boundaries of accelerated computing.