On-Device ML Infrastructure Engineer (ML User Experience APIs), Graphics, Games and Machine Learning

Experteer Overview In this role you will help shape Apple’s ML infrastructure by building model conversion and authoring APIs for end-to-end model deployment on Apple platforms. You will demonstrate and optimize cross-ecosystem workflows, integrating models from external repositories like Hugging Face with Apple’s stack. You’ll design and stress-test optimizations across source-level code and Apple representations to achieve strong performance on diverse hardware. This is a hands-on, impact-driven position at the intersection of research, software, and hardware engineering, focused on delivering a premier on-device ML experience. Compensation / Benefits * Develop and expose ML model conversion and authoring APIs as the main entry point into Apple’s ML infrastructure * Onboard popular ML models via end-to-end workflows highlighting authoring and runtime capabilities * Integrate Apple ML tools into internal and external model repositories (e.g., Hugging Face) * Ideate, design and stress test optimizations from PyTorch programs to custom transformations in Apple’s model representation * Support engineering of end-to-end inference stack from model creation to deployment Tasks * Confirmed understanding of ML modeling (architectures, training vs. inference trade-offs) * Experience in ML deployment optimizations (quantization) * Strong Python API design experience * Proficiency in Python and familiarity with C++ * Experience with ML authoring frameworks (e.g., PyTorch, MLX, JAX) * Experience with MLIR/LLVM or similar compiler toolchains * Familiarity with Hugging Face or other model repositories * Bachelor in Computer Science, Engineering, or related subject area * Hands-on experience with ML inference optimizations (quantization, pruning, KV caching) * Strong communication skills and ability to engage multi-functional audiences Key requirements *

Experteer Overview In this role you will help shape Apple’s ML infrastructure by building model conversion and authoring APIs for end-to-end model deployment on Apple platforms. You will demonstrate and optimize cross-ecosystem workflows, integrating models from external repositories like Hugging Face with Apple’s stack. You’ll design and stress-test optimizations across source-level code and Apple representations to achieve strong performance on diverse hardware. This is a hands-on, impact-driven position at the intersection of research, software, and hardware engineering, focused on delivering a premier on-device ML experience. Compensation / Benefits * Develop and expose ML model conversion and authoring APIs as the main entry point into Apple’s ML infrastructure * Onboard popular ML models via end-to-end workflows highlighting authoring and runtime capabilities * Integrate Apple ML tools into internal and external model repositories (e.g., Hugging Face) * Ideate, design and stress test optimizations from PyTorch programs to custom transformations in Apple’s model representation * Support engineering of end-to-end inference stack from model creation to deployment Tasks * Confirmed understanding of ML modeling (architectures, training vs. inference trade-offs) * Experience in ML deployment optimizations (quantization) * Strong Python API design experience * Proficiency in Python and familiarity with C++ * Experience with ML authoring frameworks (e.g., PyTorch, MLX, JAX) * Experience with MLIR/LLVM or similar compiler toolchains * Familiarity with Hugging Face or other model repositories * Bachelor in Computer Science, Engineering, or related subject area * Hands-on experience with ML inference optimizations (quantization, pruning, KV caching) * Strong communication skills and ability to engage multi-functional audiences Key requirements *

On-Device ML Infrastructure Engineer (ML User Experience APIs), Graphics, Games and Machine Learning