2 days ago
Bellevue, WA, USASenior
Base Salary
$146k - $218k/yr
Responsibilities
- Build performance and power models connecting GPU microarchitecture, software execution, and AI workload behavior.
- Characterize training and inference workloads to identify bottlenecks, validate hardware features, and inform silicon roadmap decisions.
- Collaborate with GPU hardware and firmware teams on memory subsystems, cache coherence, execution units, and next-generation Instinct designs.
- Build workload-characterization methodology, tooling, benchmarks, and performance regression infrastructure.
- Contribute to IREE compiler and ROCm runtime optimizations, runtime API extensions, firmware command packets, and GPU dispatch paths.
- Write and optimize GPU kernels and compiler passes for AMD GPU microarchitectures.
- Improve Linux kernel GPU compute infrastructure, asynchronous execution, data movement, pipelining, and latency.
- Represent software considerations in silicon architecture reviews and hardware realities to software and framework teams.
Requirements
- Bachelor’s or master’s degree in Computer Science, Computer Engineering, or a related field.
- Industry experience at the boundary of GPU compute hardware and AI systems software is preferred.
- Strong foundation in computer architecture, including memory consistency, coherence, GPU microarchitecture, and pipelining.
- Demonstrated ability to build analytical or simulation-based performance models and use them for engineering decisions.
- Proficiency in C and C++ for system-level work and Python for tooling, analysis, and framework-level development.
- Experience writing and optimizing GPU compute kernels using CUDA and/or HIP.
- Familiarity with Linux kernel internals for GPU compute, including DRM/GEM, kernel-mode driver interfaces, memory management, and userspace-kernel interactions.
- Experience with compute runtimes and dispatch paths from AI frameworks to GPU execution.
- Experience with AI framework internals such as the PyTorch dispatcher and Triton compiler.
- Direct experience with AMD Instinct architectures such as MI300X, MI350X, or MI355X and the ROCm stack is preferred.
- Working knowledge of LLVM/MLIR compiler flows, backend code generation, or optimization passes; IREE experience is a strong plus.
- Familiarity with modern AI model architectures, GPU performance analysis tools, profilers, trace viewers, hardware counters, and custom tooling.
- Publications or patents in HPC, ML systems, or GPU kernel optimization are preferred.
- Strong communication, presentation, collaboration, and technical knowledge-sharing skills.
Benefits
- The position is located in San Jose, California.
- AMD benefits are described through its benefits-at-a-glance information.
About AMD
AMD designs and sells CPUs, GPUs, and adaptive/embedded computing products for PCs, data centers, gaming, and edge devices. Its portfolio includes Ryzen and EPYC processors, Radeon and Instinct graphics, and adaptive SoCs from its Xilinx acquisition, sold to OEMs, cloud providers, and device makers. Founded in 1969 and headquartered in Santa Clara, it is a public company on NASDAQ and supplies semi-custom chips for major game consoles.
