The AIML Multimodal Foundation Model Team is pioneering next-generation intelligent agent technologies that combine multimodal reasoning, tool-use, and visual understanding. Our innovative features redefine how hundreds of millions of people utilize their computers and mobile devices for search and information retrieval. Our universal search engine powers search capabilities across a range of Apple products, including Siri, Spotlight, Safari, Messages, and Lookup. Additionally, we develop cutting-edge generative AI technologies based on multimodal large language models to enable innovative features in both Apple’s devices and cloud-based services. As a member of this team, you will design new architectures for multimodal agents, explore advanced training paradigms, and build robust agentic capabilities such as planning, grounding, tool-use, and autonomous task execution. You will collaborate closely with researchers and engineers to bring cutting-edge agent research into production, transforming Apple devices into intelligent partners that help users get things done.

Developing state-of-the-art multimodal foundation models for Apple Intelligence.

Developing various agent capabilities for multimodal LLMs, including computer use agents, visual tool use, thinking with images, and multimodal web search.

Developing, fine-tuning, and evaluating domain specific foundation models for various tasks and applications in Apple’s AI powered products

Conducting applied research to transfer the pioneering research in generative AI to production ready technologies

Understanding product requirements, translate them into modeling tasks and engineering tasks