Lead AI/ML Compiler Engineer for High-Performance GPUs

AMD; San Jose (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

AMD is seeking a highly motivated PMTS AI/ML Compiler Engineer to join its AI Software organization in California. You will drive compiler transformations and performance optimizations for AI training and inference on AMD GPUs, interfacing with architecture, kernel, runtime, and framework teams to push the ROCm/ONNX runtimes forward.

You will design scalable, portable solutions, implement and validate ONNX operators, lead performance initiatives, and analyze workload characteristics to improve

Qualifications

• Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent.
• Strong C++ and GPU programming fundamentals with experience optimizing AI workloads.
• Experience with MLIR/LLVM and performance-focused development.

Responsibilities

• Improve future AMD products with architecture teams.
• Design, develop, and optimize compiler transformations within MLIR, LLVM, and the ROCm software stack.
• Develop portable, high-performance solutions across AMD GPU architectures.
• Collaborate with compiler and runtime teams on AI workload optimization.
• Implement, validate, and maintain ONNX operators and microbenchmark test suites.
• Lead performance optimization for AI training and inference on AMD GPUs.
• Analyze workload characteristics to improve compute efficiency and memory usage.
• Drive performance benchmarking in pre- and post-silicon environments.

Skills

C++

GPU programming

Performance optimization

Software debugging

Education

Bachelor’s or Master’s degree in Computer Science/Engineering

Tools

HIP

CUDA

Triton

MLIR

LLVM

Git

Job description

AMD is seeking a highly motivated PMTS AI/ML Compiler Engineer to join its AI Software organization in California. You will drive compiler transformations and performance optimizations for AI training and inference on AMD GPUs, interfacing with architecture, kernel, runtime, and framework teams to push the ROCm/ONNX runtimes forward.

You will design scalable, portable solutions, implement and validate ONNX operators, lead performance initiatives, and analyze workload characteristics to improve