On-Device Machine Learning Engineering Manager - Tools & Services

Summary

The On-Device Machine Learning team at Apple is responsible for enabling the Research to Production lifecycle of cutting-edge machine learning models that power magical user experiences on Apple's hardware and software platforms. The team builds critical infrastructure that begins with onboarding the latest machine learning architectures to Apple devices, optimization toolkits to optimize these models to better suit the target devices, machine learning compilers and runtimes to execute these models as efficiently as possible, and the benchmarking, analysis and debugging toolchain needed to improve on new model iterations. This infrastructure underpins most of Apple's critical machine learning workflows across Camera, Siri, Health, Vision, etc., and as such is an integral part of Apple Intelligence. Our group is seeking an Engineering Manager to lead the Performance Tools and Services team, with a focus on the tools, services, and infrastructure that make on-device ML performance measurable, understandable, and improvable. The team is responsible for the frontend web services for introspecting ML models and their on-device execution, the backend web services that power them, and the on-device toolchain that gathers low-level performance data, associates it with high-level (PyTorch) framework-level ops, and reports it. The team is also responsible for the infrastructure for running ML inference across fleets of devices.

Minimum Qualifications

BS/MS/PhD in Computer Science or Electrical Engineering. Two or more years of strong and validated management experience. Knowledge of ML development and workflows, including at least one authoring framework experience (e.g., PyTorch). Experience delivering web services and/or developer-facing tools, spanning frontend and backend. Excellent communication skills. Track record of creating clean software architectures, intuitive designs, and high-performance extensible software. Solid programming skills in at least one of the following languages: Python, Swift, Objective-C, C/C++.

Preferred Qualifications

Experience with on-device ML frameworks (Core ML, Win ML, ONNX, TF Lite or ExecuTorch). Experience with performance profiling, benchmarking, and analysis tooling. Experience building and operating infrastructure for running workloads across fleets of devices. Experience associating low-level performance data with framework-level operations. Experience with MLIR / LLVM compiler technologies.

Description

We are building the first end-to-end developer experience for ML development that, by taking advantage of Apple's vertical integration, allows developers to iterate on model authoring, optimization, transformation, execution, debugging, profiling and analysis. This role focuses on giving ML developers fast, accurate, and actionable insight into how their models execute on Apple devices. We're looking for a manager that has proven experience in and passion for providing high quality developer tools and capabilities in the fast paced and dynamic space of ML. As the manager in this role, you will lead a diverse team spanning full-stack web development, backend services, distributed systems, and low-level on-device performance tooling. You will partner with leaders across the organization and company to develop our platform while supporting clients internally and externally. The role requires a solid technical understanding of ML execution on device, performance analysis and profiling, and the systems that connect low-level signals to framework-level (e.g., PyTorch) semantics.

Responsibilities

Leading and growing the team that builds the frontend web services for introspecting ML models and their on-device execution, the backend services that support them, and the on-device toolchain for gathering and reporting low-level performance data. Driving future efforts to improve the “model authoring → performance profiling → improve model → repeat“ workflow, making it faster and more insightful for ML developers. Owning the infrastructure for running ML inference across fleets of devices, ensuring it is reliable, scalable, and produces trustworthy performance data. Attracting, hiring, and guiding the career of talented ML, software, and tools engineers. Working with cross functional partners to push the state-of-the-art of on-device ML functionality and performance.