Machine Learning Architect, Platform Architecture

Summary

At Apple, our Platform Architecture group is responsible for connecting our hardware and software into one unified system. You’ll collaborate with engineers across Apple to design how our technologies work in unison, drive development of our renowned system-on-a-chip architecture and forward-looking prototype systems. Our team works at the intersection of ML applications and Apple silicon architecture. We collaborate with SoC/IP architecture, system, software, and algorithm teams to develop integrated, highly optimized solutions for machine learning applications.

About the role

In this role, you will explore different ways of mapping ML workloads to Apple silicon and develop performance models/simulations. Your work will inform and validate architecture decisions. You will critically evaluate ML model optimization techniques from the literature, analyzing what works and why, and proposing new ideas that build on what you learn. You will gain insights on how to make workloads run efficiently on our SoCs and provide guidance to software and algorithm teams.

Responsibilities

• Create optimized implementations of ML workloads on Apple silicon including Neural Engine, GPU, and CPU.
• Collaborate with IP and SoC architecture teams to develop performance models and simulations of future hardware.
• Collaborate with system teams to create high-level performance models of emerging ML techniques and analyze system architecture trade-offs.
• Evaluate emerging ML model optimization techniques through experimentation and analysis; propose new ideas to inform hardware and algorithm direction.

Minimum Qualifications

• Bachelor’s degree
• Experience in C/C++ and/or Python
• Experience in hardware IPs: ML HW accelerators, GPU/CPU, image/video processors or similar.
• Experience with ML frameworks (e.g. PyTorch) and efficient implementations of machine learning algorithms

Preferred Qualifications

• MS or PhD in EE/CE/CS or related field
• 20+ years of relevant experience
• Experience in optimizing and deploying ML models and/or runtime frameworks in production inference/training environments
• Experience designing experiments to evaluate ML model optimization techniques
• Ability to prototype algorithms on CPU/GPU/Neural Engine, analyze performance metrics, and create high-level complexity models
• Verbal and written communication skills for collaborating with partner teams
• Understanding of compilers