Machine Learning Engineer, Siri Attention & Invocation

Apple•Cupertino, CA

23d

About The Position

The Siri Attention & Invocation team is looking for Machine Learning Engineers passionate about developing and advancing frictionless voice invocation experiences on Apple’s innovative devices, enabling compelling new conversational features for Siri interactions. Build end-to-end model training and evaluation pipelines. Deploy machine-learned, on-device models that are aligned with the core values of Apple, ensuring the highest standards of quality, innovation, and respect for user privacy. And work with the people who created the intelligent assistant that helps millions of people around the world get things done — just by saying ‘(Hey) Siri.’ You will be part of a team whose focus will be on applied machine learning, on building and deploying models that constantly advance the state-of-the-art. But that is only half the story! In Siri Attention & Invocation, we own our user journeys end-to-end. We measure the impact of our deployed models not just on pre-ship evaluation sets, but also post-ship on production traffic. We optimize error rates on existing data. We also define new metrics that take into account the user experience we want to deliver and apply them to the data that best represents the next feature we ship. And we are sometimes constrained by the limits of on-device computation — that is where your ability to innovate will be most impactful. You will collaborate with many dynamic, cross-functional teams consisting of software engineers and machine learning engineers/scientists. The ideal candidate will excel in both academic rigor and engineering efficacy, staying up-to-date with the latest research advancements as well as delivering reliable and robust models to all devices for all users around the world. If you are passionate about building outstanding products and using the full spectrum of your skills to extend the core technology that lets Siri understand, personalize, and interact in new and exciting ways, then we cannot wait to hear from you. “Hey Siri, let’s work together at Apple!" The Siri Attention & Invocation team is looking for Machine Learning Engineers passionate about developing and advancing frictionless voice invocation experiences on Apple’s innovative devices, enabling compelling new conversational features for Siri interactions. Build end-to-end model training and evaluation pipelines. Deploy machine-learned, on-device models that are aligned with the core values of Apple, ensuring the highest standards of quality, innovation, and respect for user privacy. And work with the people who created the intelligent assistant that helps millions of people around the world get things done — just by saying ‘(Hey) Siri.’

Requirements

3-5 years of experience with scalable machine learning technologies
Strong background in machine learning and deep learning
Proficiency in deep learning / machine learning frameworks (e.g., PyTorch, TensorFlow)
Proficiency in programming languages including but not limited to C/C++/Python
Strong software engineering fundamentals
Interest in optimizing, automating, and scaling end-to-end systems (e.g., PySpark, Airflow)
Strong attention to detail
Analytical skills
Willingness to dive into data to explain anomalies and conduct error/deviation analyses
Outstanding problem solving, critical thinking, creativity, and interpersonal skills
Ability to communicate effectively
Ability to work well in multi-functional teams

Nice To Haves

Experience in speech recognition is highly desired
Master’s or Ph.D. degree in Computer Science, Electrical Engineering or related field
Outstanding candidates with Bachelor’s degrees and multiple years of significant engineering/product experience will also be considered
Industry experience in product development and deployment
Understanding of full software product life cycle

Responsibilities

Build end-to-end model training and evaluation pipelines.
Deploy machine-learned, on-device models that are aligned with the core values of Apple, ensuring the highest standards of quality, innovation, and respect for user privacy.
Work with the people who created the intelligent assistant that helps millions of people around the world get things done — just by saying ‘(Hey) Siri.’
Focus on applied machine learning, on building and deploying models that constantly advance the state-of-the-art.
Own user journeys end-to-end.
Measure the impact of deployed models not just on pre-ship evaluation sets, but also post-ship on production traffic.
Optimize error rates on existing data.
Define new metrics that take into account the user experience we want to deliver and apply them to the data that best represents the next feature we ship.
Innovate within the limits of on-device computation.
Collaborate with many dynamic, cross-functional teams consisting of software engineers and machine learning engineers/scientists.
Stay up-to-date with the latest research advancements.
Deliver reliable and robust models to all devices for all users around the world.
Extend the core technology that lets Siri understand, personalize, and interact in new and exciting ways.