Machine Learning System Software Engineer

Quick Facts

• Role: Machine Learning System Software Engineer (Neural Engine)
• Focus: System software APIs, inference runtime design, performance optimization

Description

You will design and implement APIs for the Apple Neural Engine that define how models are loaded, compiled, scheduled, and executed on device. The role involves creating runtime APIs for lifecycle management, memory allocation strategies, multi-model scheduling, and concurrent execution, then profiling and optimizing end-to-end inference performance. As the position evolves, you will take on technical leadership and collaborate across driver, firmware, and compiler teams.

Responsibilities

• Design and implement Neural Engine APIs for client model loading, compilation, scheduling, and execution
• Architect APIs for model lifecycle management, memory allocation strategies, multi-model scheduling, and concurrent execution
• Co-design the runtime-to-silicon interface with driver, firmware, and compiler teams
• Profile and optimize scheduling, compilation, and inference execution paths to remove bottlenecks
• Support clients in adopting the APIs and integrating efficiently

Requirements

• BS degree and at least 3 years of relevant industry experience
• Experience defining interfaces used by other teams/external developers with lifecycle, error handling, and forward compatibility
• Deep proficiency in C and C++ in large production system software
• Understanding of runtime systems: process/thread models, memory management, IPC/RPC, and resource lifecycle
• Understanding of software–hardware interfaces: registers, DMA, command queues, or similar accelerator interaction patterns
• Experience shipping production system software

Benefits

• Opportunities to demonstrate technical leadership and influence key design decisions
• Collaborative environment with cross-team work across hardware and software domains
• Chance to help shape AI-driven capabilities across an ecosystem of Apple products