Staff Software Development Engineer - Rust, Compilers, and GPU Systems (1)
Build low-level GPU systems software in Rust, shaping compiler pipelines, runtimes, and firmware for AMD’s next-gen hardware while ensuring memory safety and correctness across the stack.
- Compensation
- Not disclosed
- Location
- US,CA,San Jose
- Work setup
- Not specified
- Seniority
- Staff
- Employment
- Not specified
- Experience
- Not specified
- Category
- Software engineering
- Posted
Skills
Nice to have
- Strong production Rust experience, including ownership and borrowing, lifetimes, traits and generics, error handling, concurrency, FFI, and the careful design of safe abstractions over unsafe operatio
- Experience with low-level or constrained Rust, including `no_std`, embedded systems, firmware, custom targets, allocators, volatile access, interrupt handling, or architecture-specific code.
- Strong C or C++ systems programming skills and the ability to work across mixed-language codebases.
- Experience with rustc internals, MIR, custom codegen backends, procedural macros, compiler plugins, or other Rust toolchain infrastructure.
- Experience with LLVM, MLIR, GCC, or another production compiler framework. Experience designing intermediate representations, dialects, analyses, optimization passes, legalization, instruction selecti
- Understanding of compiler correctness issues such as undefined behavior, aliasing, provenance, layout, calling conventions, memory models, and source-to-target semantic preservation.
- Ability to read and reason about low-level assembly, ISA encodings, compiler output, and machine-level execution.
- Experience with linkers, loaders, ELF or other object formats, code-object metadata, DWARF, LLVM MC, disassembly, post-link analysis, or binary rewriting.
- Understanding of GPU execution models, including waves or warps, SIMD/SIMT execution, address spaces, registers, local or shared memory, caches, synchronization, atomics, barriers, and asynchronous op
- AMDGPU, GCN, RDNA, CDNA, ROCm, HIP, HSA, or AMDGPU LLVM backend experience.
- Experience with GPU runtimes, drivers, firmware, embedded processors, device initialization, command processing, queueing, or hardware control paths.
- Experience building secure or high-assurance systems that process untrusted inputs or operate across privilege, memory-isolation, or hardware trust boundaries.
- Experience with program analysis, abstract interpretation, symbolic execution, model checking, SMT solvers, or formal verification tools such as Verus, Lean, Coq, Isabelle, Dafny, or TLA+.
- Experience writing executable specifications, proving state-machine invariants, or validating refinement between source, IR, binaries, simulators, RTL, and hardware.
- Strong validation instincts, including adversarial testing, differential testing, property-based testing, fuzzing, reproducible evidence, and careful treatment of assumptions and counterexamples.
- Hardware-software co-design experience for ISA features, compiler-visible architecture, memory systems, synchronization, security mechanisms, or specialized compute pipelines.
- Technical leadership experience: setting direction, resolving ambiguous cross-layer problems, influencing partner teams, mentoring engineers, and communicating tradeoffs clearly.
Full description
ADVANCE YOUR CAREER. ADVANCE THE WORLD.
At AMD, we believe technology can change lives for the better. It can heal us, entertain us, and make us more connected, productive, and understanding of the world around us. And we’re looking for talent who feel the same: people who want to leave the planet better than they found it, those who don’t shy away from humanity’s challenges but are determined to help solve them.
AMD is powering the next generation of supercomputing, high-performance computing, cloud, and AI. Whether you’re designing next-gen processors, enabling AI breakthroughs, or creating go-to-market plans, every role at AMD contributes to something bigger — technology that moves the world forward.
THE ROLE:
We are building next-generation systems software for AMD GPUs with Rust as a core technical direction. The work spans compilers, runtimes, low-level GPU software, firmware, developer tooling, and methods for improving the safety and correctness of complex hardware-software systems.
Rust is not incidental to this role. You will help establish how Rust is used in performance-sensitive and high-trust parts of the GPU stack, including the compiler and tooling support, system interfaces, engineering practices, and validation methods required for production deployment. You will have the
opportunity to influence both near-term implementations and the longer-term architecture for using Rust across current and future AMD platforms.
THE PERSON:
This role is for a deeply technical compiler and systems generalist who enjoys working across abstraction layers. You may move between Rust source semantics, rustc internals, intermediate representations, MLIR and LLVM, target-specific code generation, object formats, runtimes, GPU ISA, firmware, and hardware interfaces. You should have strong judgment about correctness boundaries. Memory safety is important, but production correctness also depends on concurrency, ordering, ABI compatibility, device behavior, compiler transformations, and the assumptions between software and hardware. You will help make those assumptions explicit and build evidence that the resulting systems behave as intended.
The ideal candidate has deep expertise in at least one part of this stack and the ability to become productive in unfamiliar areas. Experience with every technology listed below is not expected. A research background is not required, although the ability to read technical work, form testable
hypotheses, and turn promising ideas into reliable implementations is valuable. The work sits at the intersection of Rust systems programming, production compiler infrastructure, GPU architecture, firmware, program analysis, and formal methods. Hardware-software co-design experience is highly valuable, especially when compiler, runtime, firmware, or ISA choices must evolve