Machine Learning Engineer Applied Science

Siri helps hundreds of millions of people find the information they are looking for. A critical part of that mission is helping them quickly find and discover local businesses places of interest and addresses. Users rely on us for relevant and easy access to local information like finding a favorite or romantic restaurant business hours nearby coffee shop addresses and directions to prominent locations. The Geo domain team is redefining how hundreds of millions of people use their devices to navigate and explore the physical world around them. We are part of a wider effort to power search across a variety of Apple products including Siri Spotlight Safari Messages and more. As part of our team you will be using innovative machine learning techniques and LLMs in order to understand queries rank documents and find useful answers to users questions. We are looking for an experienced applied researcher with hands-on experience in search and recommendation and deploying powerful machine learning models at scale. You will join a team that combines strong technical skills product vision and a love of all things local to bring together the pieces needed to deliver an extraordinary Maps experience in Siri and Spotlight.

As a member of our high-impact iterative environment youll have the unique and rewarding opportunity to shape upcoming products from Apple. Our team includes a diversity of backgrounds from applied scientists with a focus in NLP to experienced distributed systems engineers. We are looking for candidates with both applied machine learning and deep-learning experience as well as strong engineering skills.

Own the entire ML development cycle from opportunity analysis exploration and prototyping to data collection feature engineering training evaluation and deployment in the development of machine learning models to improve search quality across retrieval ranking reranking and query search quality and experience by leveraging techniques such as learning-to-rank embedding models contrastive learning multi-task learning and reinforcement learning where identify high-impact research directions and drive them to production translating state-of-the-art findings into measurable improvements in search cross-functionally with product and design teams to shape the technical roadmap and translate research capabilities into user-facing junior engineers and provide technical leadership in architecting ML systems and designing ML product requirements then drive the technical design and model architecture before defining the roadmap. Own the final outcome and learn on every iteration.

You have 8 years of experience in information retrieval natural language processing machine learning or deep have a deep understanding of machine learning theory including supervised learning ranking models embeddings representation learning and evaluation have proven ability to apply advanced ML techniques to improve search relevance and retrieval quality at are comfortable leading experimentation offline evaluation and online A/B testing for iterative improvements in search actively monitor recent research literature including arXiv NeurIPS ICML ACL SIGIR and industry publications and have a track record of translating findings into practical system independently identify high-impact research directions and drive them forward without requiring top-down demonstrate a strong bias toward action moving fluidly from paper to prototype to production in tight iteration cycles executing quickly while maintaining quality and have excellent interpersonal skills the ability to work independently as well as part of a team including cross-functional collaboration with product and have a Masters Degree in Computer Science Machine Learning or a related field or equivalent practical experience.

PhD in Computer Science Machine Learning Information Retrieval or a related field or equivalent research experience demonstrated through publications patents or significant open-source record of publishing or presenting at top-tier research venues such as NeurIPS ICML ACL SIGIR WWW or applying LLMs and generative AI techniques to production search or recommendation systems.