Machine Learning Engineer - Notifications & Personalization

Apple•Cupertino, CA

About The Position

Our team is looking for you to help make iOS more intelligent, proactive and personal. Our team is part of the core iOS experience, using privacy preserving on-device intelligence to drive new experiences that touch the lives of millions of Apple customers every day. We are responsible for personalizing core system experiences, such as helping you manage and summarize notifications, get the most relevant widgets in smart stacks, as well as predicting your next Focus. In our team you will bring expertise in machine learning and software engineering to train and deploy models that surprise and delight our customers every day. You will work closely with talented Software and ML engineers on our team, and across Apple to train, deploy, and evaluate machine learned models that power new experiences across iOS and all Apple platforms. As we build the future of iOS, you will be responsible for driving the development of the models and infrastructure that power intelligent system experiences. You will be responsible for providing technical leadership across a wide variety of products and features, proposing and executing innovative solutions, and improving our model training and evaluation pipelines in service of quality and future feature development. You will be a core participant in developing new models and architectures that result in high quality features while adhering to device power and performance constraints. We are passionate about user experience and privacy. Our mission is to craft user experiences which leverage the power of machine learning and on-device intelligence to preserve our customers privacy. Come help build state-of-the art intelligence impacting over a billion users.

Requirements

Expertise in machine learning and software engineering.
Experience shipping and evaluating at scale.
Experience shipping ML models to millions of users and interpreting real-world metrics to evaluate model and product impact.
Designing and running evaluation frameworks to measure model quality, user engagement, and product outcomes at scale.
Building instrumentation and telemetry pipelines to collect, analyze, and act on user signals across large populations.
Planning and implementation of experimentation of various on-device machine learning models to understand the performance of deployed models and optimize performance.
Experience training and deploying machine learned models.
Implementing, training, and optimization of machine learning models with the application of unsupervised and supervised learning techniques including classification, regression, and artificial neural network algorithms using open source libraries such as Scikit-Learn and Apple-owned frameworks such as CoreML and CreateML.
Applying reinforcement learning techniques to personalize on-device suggestions including app, people, and action recommendations.
Deploying machine learning models using Python, Objective-C, and Swift for on-device inference, applying software engineering skills to write clean, readable, testable, and deployable code.
Identifying and selecting appropriate datasets and data representation methods, including identification of data sources and implementation of data collection algorithms to collect privacy-preserved data for model training.
Applying data preparation techniques including preprocessing, profiling, cleansing, validation, and transformation to prepare data for model training.
Applying statistical and mathematical skills including regression analysis, using tools such as Python and SQL-based querying tools to derive insights from large data sets for optimization of model performance.
Strong product and design intuition — ability to reason about how model decisions surface in the UI and affect user perception.
Collaborating closely with the Apple Design team to ensure intelligent features feel intentional, not intrusive.
Translating ambiguous user needs into measurable signals and tuning models to optimize for user-perceived quality rather than purely statistical metrics.
Understanding of large-scale operating systems development and how to build within constraints like performance, memory, and power.
Experience building across the full iOS system stack using Objective-C, Swift, and C++.
Experience building features that span multiple system components, with familiarity in on-device inference constraints including latency, privacy, and resource budgets.
Cross-functional collaboration with teams such as Privacy Engineering to ensure privacy-preserved and secured data collection for model training and optimization.
Experience with UI development, design and prototyping.
7-10 years of experience: Shipping and evaluating at scale, Training and deploying machine learned models, Building great user experiences, Working on embedded operating systems.
BS, M.S. or PhD in Software Engineering, Computer Science, Machine Learning, or related field.

Responsibilities

Driving the development of the models and infrastructure that power intelligent system experiences.
Providing technical leadership across a wide variety of products and features.
Proposing and executing innovative solutions.
Improving our model training and evaluation pipelines in service of quality and future feature development.
Developing new models and architectures that result in high quality features while adhering to device power and performance constraints.
Shipping and evaluating ML models to millions of users and interpreting real-world metrics to evaluate model and product impact.
Designing and running evaluation frameworks to measure model quality, user engagement, and product outcomes at scale.
Building instrumentation and telemetry pipelines to collect, analyze, and act on user signals across large populations.
Planning and implementation of experimentation of various on-device machine learning models to understand the performance of deployed models and optimize performance.
Implementing, training, and optimization of machine learning models with the application of unsupervised and supervised learning techniques including classification, regression, and artificial neural network algorithms using open source libraries such as Scikit-Learn and Apple-owned frameworks such as CoreML and CreateML.
Applying reinforcement learning techniques to personalize on-device suggestions including app, people, and action recommendations.
Deploying machine learning models using Python, Objective-C, and Swift for on-device inference, applying software engineering skills to write clean, readable, testable, and deployable code.
Identifying and selecting appropriate datasets and data representation methods, including identification of data sources and implementation of data collection algorithms to collect privacy-preserved data for model training.
Applying data preparation techniques including preprocessing, profiling, cleansing, validation, and transformation to prepare data for model training.
Applying statistical and mathematical skills including regression analysis, using tools such as Python and SQL-based querying tools to derive insights from large data sets for optimization of model performance.
Translating ambiguous user needs into measurable signals and tuning models to optimize for user-perceived quality rather than purely statistical metrics.
Building across the full iOS system stack using Objective-C, Swift, and C++.
Building features that span multiple system components, with familiarity in on-device inference constraints including latency, privacy, and resource budgets.
Cross-functional collaboration with teams such as Privacy Engineering to ensure privacy-preserved and secured data collection for model training and optimization.
Experience with UI development, design and prototyping.