Edge AI Compiler Engineer for Custom Silicon

Meta; Redmond; On-site; AUD 262,000 - 366,000; Full time

Job summary

Reality Labs at Meta is hiring a Software Engineer for compiler tool-chain development focused on state-of-the-art DL hardware accelerators. You will help design and implement an SL-internal compiler stack and build LLVM/MLIR-based tools for PyTorch model deployment to custom silicon.

You will work with hardware architects and ML engineers to optimize performance, power efficiency, and edge suitability across fabrics, GPUs, and TPUs.

Qualifications

• 3+ years of experience in developing compilers, toolchains, or code optimization software.
• Experience with LLVM, MLIR, or similar compiler infrastructure frameworks.
• Experience designing intermediate representations and implementing compiler optimization passes.
• Experience with GPUs, TPUs, or custom AI accelerators.
• Strong software development in C++ for compiler and systems-level programming.
• Experience leading end-to-end compiler infrastructure initiatives across teams.

Responsibilities

• Lead the architecture and implementation of ML compiler infrastructure, including IRs, optimizations, and code generation for custom accelerators.
• Design compiler transformations informed by hardware constraints for GPU, TPU, and edge AI targets.
• Advance LLVM/MLIR toolchains to compile PyTorch models to optimized binaries for custom silicon.
• Collaborate with hardware architects to maximize performance, power efficiency, and programmability.
• Improve efficiency, scalability, and stability of compiler toolchains across targets.
• Lead roadmapping for compiler infrastructure and mentor engineers in design patterns.
• Review designs and code; address cross-disciplinary system issues.
• Interface with PyTorch and ExecuTorch teams to evaluate innovations.

Skills

Compiler development

C++

IR design

LLVM/MLIR

Cross-team leadership

Education

Bachelor's degree

Tools

LLVM/MLIR

PyTorch

TensorFlow

ExecuTorch

Job description

Reality Labs at Meta is hiring a Software Engineer for compiler tool-chain development focused on state-of-the-art DL hardware accelerators. You will help design and implement an SL-internal compiler stack and build LLVM/MLIR-based tools for PyTorch model deployment to custom silicon.

You will work with hardware architects and ML engineers to optimize performance, power efficiency, and edge suitability across fabrics, GPUs, and TPUs.