On-device ML Infrastructure Engineer (Orchestration & Performance)

What you'll do

• Drive full-stack changes across the OS and tooling to deploy large SOTA models on Apple Silicon.
• Implement mechanisms to support efficient orchestration of machine learning models across various hardware devices.
• Modify authoring tools and MLIR-based compilers to expose mechanisms for state-of-the-art model execution.
• Develop infrastructure to enable high-performance, stable execution of models across the Apple ecosystem.
• Build optimization toolkits for model compression and acceleration on embedded systems.
• Create comprehensive benchmarking and debugging toolchains for machine learning workflows.
• Develop ML compilers and runtimes to ensure efficient execution across diverse hardware.

What we're looking for

• 3-5 years of experience working on tooling built in Python 3 and C++/Swift.
• Familiarity with common ML model architectures, execution schemes, and operations.
• Familiarity with PyTorch or related training frameworks.
• Experience working on or adjacent to MLIR-based compilers.
• Familiarity with deploying applications or tooling on Apple platforms.
• Familiarity with programming paradigms for the GPU, CPU, and Neural Engine.
• Familiarity with writing kernels for ML model execution.