On-Device ML Performance Engineer

About the role

## About the Company

Apple is a global technology company known for designing innovative hardware, software, and services. Its On-Device Machine Learning team works across research, software engineering, hardware engineering, and product teams to bring advanced machine learning capabilities to Apple devices.

## About the Role

Apple is seeking an **On-Device ML Performance Engineer** to analyze and optimize the performance of machine learning models running on the latest iPhone and Mac hardware.

In this role, you will work across the software and hardware stack, analyzing latency, memory usage, power consumption, and numerical correctness. You will help ensure that Apple's machine learning software stack fully utilizes Apple Silicon and its CPU, GPU, Neural Engine, memory, and other hardware capabilities.

The role combines machine learning, computer architecture, performance engineering, low-level software, and hardware debugging.

## Key Responsibilities

• Analyze and optimize machine learning inference performance on Apple devices.
• Evaluate ML models created using frameworks such as PyTorch and MLX.
• Analyze latency, memory, power, and numerical correctness of on-device inference.
• Investigate performance across CPU, GPU, Apple Neural Engine, system memory, and power systems.
• Identify opportunities to improve ML model performance and efficiency.
• Support optimization techniques such as quantization and sparsity.
• Evaluate performance and accuracy trade-offs.
• Develop scripts, utilities, and debugging tools for performance analysis.
• Extract, analyze, and report performance and power metrics.
• Work closely with research, software engineering, hardware engineering, and product teams.
• Analyze the interaction between ML models, compilers, drivers, and hardware.
• Help optimize Apple's ML software stack to take full advantage of Apple Silicon.
• Contribute to performance analysis and optimization across new Apple hardware and ML inference workloads.

## Technical Expertise

The ideal candidate should have strong knowledge or experience in:

• Machine learning model architectures and inference.
• On-device machine learning.
• CPU and GPU architecture.
• Computer architecture and memory systems.
• ML performance optimization.
• Compilers and hardware drivers.
• Performance and power analysis.
• Machine learning frameworks such as PyTorch or MLX.
• Scripting and software development.
• Low-level debugging and performance analysis.
• Apple Silicon and ML accelerator technologies.

## What You Will Work On

This position provides the opportunity to work on cutting-edge machine learning models running directly on Apple hardware. You will analyze the complete inference path, from the ML software stack and low-level drivers to hardware behavior.

Your work will help improve the performance, efficiency, and accuracy of machine learning experiences across Apple's latest devices.

## Equal Opportunity

Apple is committed to providing an inclusive workplace and equal employment opportunities. Qualified applicants are considered based on their skills, qualifications, and experience without discrimination based on legally protected characteristics.