harshmenon
I'm looking for a strong compiler and systems generalist to join me at AMD in San Jose.
This role is for someone who enjoys moving across abstraction layers and solving difficult, open-ended engineering problems. Relevant experience may include LLVM, MLIR, compiler frontends or backends, intermediate representations, optimization and code generation, language runtimes, Rust or C++, low-level GPU systems, firmware, program analysis, or formal methods.
Rust is central to the role. We are investing in Rust for low-level systems where performance, control, and correctness all matter, and this person will help make that investment successful. You should be excited to write substantial production Rust and to improve the compiler, tooling, interfaces, and engineering practices around it. Deep Rust experience is valuable, but a strong compiler or C++ systems engineer who is serious about becoming highly effective in Rust can also be a good fit.
You do not need experience in every area. Strong fundamentals, sound technical judgment, and the ability to become productive in unfamiliar parts of a large systems stack matter more than matching a fixed list of technologies. We value people who can connect architecture to implementation, build rigorous tests, and communicate clearly about correctness and tradeoffs.
This is a full-time, onsite position in San Jose, California. To learn more, email me directly at [email protected] with "HN" in the subject. Please include a short introduction and a resume or links to relevant work.
Role: GPU Performance Software Development Engineer
Location: San Jose, CA (see posting for details)
We’re building Wave, a high-performance GPU programming language and compiler for machine learning workloads. Wave combines a Python-embedded DSL with an MLIR-based compiler stack to give engineers explicit control over GPU kernel performance.
We’re looking for a low-level GPU performance engineer to own end-to-end performance of Wave’s kernels.
You’ll work on
• Hand-tuned GPU kernels (GEMM, Attention, MoE, decoding)
• Instruction-level optimization: memory, registers, scheduling, wave/warp behavior
• HIP/CUDA, GPU intrinsics (MFMA/MMA), and inline assembly • MLIR dialects, lowering pipelines, and performance-critical compiler passes
• Profiling via disassembly, counters, rocprof/Nsight
Wave codebase: https://github.com/iree-org/wave
Job posting: https://careers.amd.com/careers-home/jobs/77039?lang=en-us
We are building a fundamentally new programming language for authoring high performance machine learning kernels on AMD GPUs and the rest of our hardware portfolio. To enable this, we are looking for a wide variety of roles ranging from GPU performance experts, MLIR/LLVM compiler wizards, programming language specialists and machine learning for systems researchers.
We are looking for candidates with deep domain expertise in any of the areas listed above (atleast 5 YoE and contributions to major OSS projects or publications in relevant conferences)
If this sounds interesting, please reach out to me via email (harsh dot menon at amd dot com) with the title "Interested in Wave Language" and your resume along with a brief description of what role you are interested in and why we should consider you for the role.