Sr. Applied ML Engineer, Apple Services Localization Engineering

Are you passionate about building extraordinary products and embracing state-of-the-art techniques and technologies In Apple Services Localization Engineering we are leveraging Machine Learning and Machine Translation to expand our global product reach and bring our Services to more users around the world. The Apple Services Engineering team is one of the most exciting examples of Apples passion for combining art and technology. These are the people who power Apple Music the App Store Apple TV Apple Fitness Apple Podcasts Apple Books and more. And they do it on a massive scale meeting Apples high expectations with high performance to deliver a huge variety of entertainment in over 37 languages to more than 175 countries. These engineers build secure end-to-end solutions. They develop the custom software used to process all the creative work the tools that providers use to deliver that media all the server-side systems and the APIs for many Apple services. nnThanks to Apples outstanding integration of hardware software and services engineers here partner to get behind a unified vision. That vision always includes a deep dedication to strengthening Apples privacy policy one of Apples core values. Although Services are a bigger part of Apples business than ever before these teams remain nimble and cross-functional offering greater exposure to the array of opportunities here.

We build and develop the core language and machine translation models that power Localization across Services in an efficient and scalable manner and the production systems that put those models in front of users. We work on a wide spectrum of approaches including agentic workflows foundation modeling deep learning model compression and transfer learning. We also build the systems that power Apple Music lyrics translations and lyrics transliterations (phonetic pronunciation). This position spans applied modeling and the software engineering needed to ship at scale offering a unique chance to work where Localization meets state-of-the-art large-scale software this sound like you Join our team!

In This Role You Will:nnDesign build and ship machine translation and LLM-based systems that power Localization across Apple Services including Apple Music the App Store subscription services and marketing models from prototype to production: build and own the serving inference and data pipelines that run reliably at massive scale with a focus on latency throughput and ML models into Apple Services systems and APIs partnering across teams to turn business objectives into robust technical applied research and experimentation including LLM fine-tuning model compression and emerging techniques such as agentic workflows and the evaluation infrastructure used to measure translation quality and system performance and use it to guide results recommendations and trade-offs clearly to both technical and non-technical audiences including leadership and operations engineers and foster a culture of collaboration technical excellence and innovation. Where it aligns with Apples innovation standards contribute to publications and patents.n

BS/MS/PhD in a quantitative field (Computer Science Math Statistics Physics etc.) and 5 years of relevant software engineering fundamentals and proficient programming skills in Python with experience writing and maintaining production-quality -on experience with deep learning toolkits such as JAX TensorFlow or track record training or deploying large models in building or operating large-scale distributed production understanding of Deep Learning Large Language Models (LLMs) and Natural Language Processing (NLP).

PhD in a quantitative field or equivalent depth in machine translation multilingual NLP or applied LLM with machine translation and translation-quality evaluation (e.g. COMET BLEU human evaluation).nExperience optimizing model serving and inference quantization distillation or compression for low-latency high-throughput with MLOps and model-deployment building LLM-based agents including the ReAct pattern in agentic workflows.