Staff Machine Learning Performance Engineer, Siri Runtime Systems and Interaction

Apple is where individual imaginations gather together, committing to the values that lead to great work. Every new product we build, service we create, or Apple Store experience we deliver is the result of us making each other's ideas stronger. That happens because every one of us shares a belief that we can make something wonderful and share it with the world, changing lives for the better. It's the diversity of our people and their thinking that inspires the innovation that runs through everything we do. When we bring everybody in, we can do the best work of our lives. Here, you'll do more than join something - you'll add something.\\n\\nThe Siri team in the AIML group at Apple is seeking an exceptional Machine Learning Engineer to lead efforts in identifying bottlenecks and optimizing our model inference stack. In this highly collaborative role, you will be at the center of multiple initiatives to accelerate and optimize LLMs and other ML models used by Siri. This position involves consulting with multiple product teams to determine the appropriate foundation model (On Device vs Server) for their use cases and to help them achieve their accuracy and performance targets.\\n\\nYour work will directly impact Siri's performance and efficiency, enhancing the overall user experience. You will be expected to bring innovative ideas and a problem-solving mindset to tackle the unique challenges associated with optimizing complex ML models.

As a Machine Learning Performance Engineer, you will play a critical role in ensuring the efficiency and scalability of Siri's machine learning models. You will work closely with diverse teams to diagnose performance issues and develop innovative solutions that enhance model performance. Your expertise will be pivotal in shaping the future of Siri's AI capabilities.\n\n- Analyze and optimize the performance of machine learning models and systems used by Siri.\n- Develop and implement strategies for model tuning, parameter optimization, and efficient resource usage.\n- Conduct performance benchmarking and develop tooling and metrics to measure model performance in terms of compute, memory and latency.\n- Collaborate with feature and product teams to consult on modeling decisions to achieve Siri performance objectives.\n- Collaborate with hardware and software teams to integrate research findings into product implementation.

Understanding of Transformer and LLM architectures.\nStrong understanding of Operating System, Compiler and Computer Architecture fundamentals. Expertise in optimizing software for take advantage of underlying hardware architecture.\nExperience in analyzing, identifying, and optimizing performance bottlenecks.\nBachelors degree in Computer Science, Engineering, or related discipline, or 10+ equivalent industry experience

Strong plus if you have expertise in optimizing model architectures for on device inference.\nStrong plus if you have previously worked with modeling pipeline teams in model deployment and promotion pipelines.\nCreative, collaborative, and product-focused.\nExcellent communication skills\nPhD in related field