On-Device Multimodal AI Scientist for Foundation Models

Apple Inc.; Seattle (WA); On-site; USD 142,300 - 263,300; Full time

Job summary

Apple Inc. in Seattle, WA seeks an AI Research Scientist focusing on multimodal foundation models—architecture, pre-training, and distillation—for on-device intelligence. You will push novel architectures and efficient training methods that translate to real Apple experiences.

You will explore teacher–student distillation, data strategies, and energy-aware designs, with opportunities for publication and selective open-sourcing. Collaboration across hardware, software, and AI teams is required.

Qualifications

• Experience designing, implementing, and running large-scale pre-training experiments for large language models.
• Experience with LLM pre-training topics including architecture, objectives, data mixtures, tokenization, curricula, scaling, optimization.
• Proficiency with PyTorch or JAX and distributed training systems.
• Experience evaluating pretrained models on language understanding, reasoning, instruction following, or multimodal capabilities.
• Strong understanding of transformer-based architectures and scalable training.
• Master’s degree or equivalent practical experience in ML/CS.

Responsibilities

• Advance architectures, pre-training methods, and distillation techniques for multimodal foundation models.
• Experiment at scale across architecture, data strategies, and optimization.
• Translate successful ideas into foundation-model technologies for Apple products.
• Collaborate across model research, data, systems, and hardware boundaries.

Skills

LLM pretraining experiments

LLM pretraining topics

PyTorch/JAX

Multimodal evaluation

Transformer architectures

MS or equivalent experience

Education

Master’s degree or equivalent

Job description

Apple Inc. in Seattle, WA seeks an AI Research Scientist focusing on multimodal foundation models—architecture, pre-training, and distillation—for on-device intelligence. You will push novel architectures and efficient training methods that translate to real Apple experiences.

You will explore teacher–student distillation, data strategies, and energy-aware designs, with opportunities for publication and selective open-sourcing. Collaboration across hardware, software, and AI teams is required.