Staff Rust Systems Engineer, GPU Compiler

Advanced Micro Devices; San Jose (CA); Hybrid; USD 170,000 - 210,000; Full time

Benefits offered by this job

AMD benefits at a glance

Job summary

Advanced Micro Devices in San Jose, CA, seeks a deeply technical compiler and systems generalist to lead Rust components for low‑level, high‑performance GPU software. You will influence the compiler, runtime, and firmware integration, ensuring safety, correctness, and efficiency across AMDGPU targets.

You will work across rustc internals, MLIR/LLVM, and hardware interfaces, defining rigorous interfaces and validation strategies with cross‑team partners.

Qualifications

• Deep Rust production experience and safe abstraction design.
• Strong C/C++ systems programming skills across mixed codebases.
• Experience with rustc internals, MIR, or custom backends.
• Familiarity with LLVM/MLIR, GPU backends, and compiler tooling.
• Understanding of undefined behavior, memory models, and ABI rules.

Responsibilities

• Lead design and implement production Rust components for GPU systems.
• Advance Rust usage across GPU software, toolchains, and libraries.
• Design compiler pipelines: rustc, MIR, LLVM IR, MLIR dialects, AMDGPU codegen.
• Develop analyses, lowering for memory spaces, and synchronization.
• Define interfaces across compiler, host runtime, device code, and firmware.

Skills

Rust programming

Compiler development

C/C++ systems

GPU architecture

Hardware-software co-design

Concurrency

Formal methods

Reading IRs & assembly

Education

Bachelor's degree in CS/CE or related field

Advanced degree welcome

Tools

LLVM

MLIR

GCC

rustc

GPU toolchains

ELF/Code-objects

Job description

Advanced Micro Devices in San Jose, CA, seeks a deeply technical compiler and systems generalist to lead Rust components for low‑level, high‑performance GPU software. You will influence the compiler, runtime, and firmware integration, ensuring safety, correctness, and efficiency across AMDGPU targets.

You will work across rustc internals, MLIR/LLVM, and hardware interfaces, defining rigorous interfaces and validation strategies with cross‑team partners.