About the job
Our team is an applied research and engineering team responsible for developing real-time on-device Language, Computer Vision, and Machine Perception technologies across Apple products. We focus on technology research and development to deliver Apple quality, state-of-the-art experiences. Our team prides itself on innovating through the full stack, and partnering with HW, SW and ML teams to influence the sensor and silicon roadmap that brings our vision to life. We are directly responsible for the on-device optimization and deployment of the Apple Intelligence LLM and diffusion models. As a Machine Learning Engineer, you will have the opportunity to be at the forefront of technological advancements and contribute to the successful shipping and delivery of Apple intelligence. You will be responsible for implementing and delivering various optimization techniques that improve the performance of large language and diffusion models on devices. Additionally, you will collaborate with a diverse range of organizations within Apple. Your innovations will significantly impact the entire ML model lifecycle of Apple intelligence.
Responsibilities
Implementing and delivering various optimization techniques that improve the performance of large language and diffusion models on devices; Collaborating with a diverse range of organizations within Apple
Qualifications
Minimum
Software engineering skills in Python
Experience in developing large computer vision and machine learning models, particularly on the hardware-aware model optimizations
BS and a minimum of 3 years relevant industry experience
Preferred
Familiar with model compression algorithms including quantization, pruning, distillations, and experience on optimizing large diffusion models or language models
MS or PhD degree in Computer Science, or equivalent industry research experience
Experience with hardware architecture, software & hardware co-design
Leadership experience in driving large-scale projects in the industry
Strong communication skills; phenomenal work ethic and collaboration
ML compiler
High performance kernel implementation
Distributed inference