AI Research Scientist: Multimodal Foundation Models - Architecture, Pre-Training & Distillation

What you'll do

• Develop and evaluate architectures, pre-training objectives, data strategies, optimization methods, and teacher–student learning techniques.
• Run research experiments on foundation-model training, scaling, distillation, and evaluation.
• Co-develop frontier and efficient models for Apple silicon and on-device intelligence.

What they ask for

• Hands-on 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, and optimization.
• Proficiency with PyTorch or JAX and distributed training systems; experience evaluating pre-trained models.
• Strong understanding of transformer-based architectures and efficient or scalable foundation-model training.

Nice to have

• Experience with knowledge distillation or teacher–student training pipelines.
• Experience with multimodal models, inference efficiency, hardware–software co-design, or relevant research contributions.

Pay

$142,300 and $263,300 USD (base salary); $150,400 and $277,600 USD (base salary)

Tools

PyTorch, JAX

AI focus

Research and develop multimodal foundation models, including architecture, pre-training, scaling, and distillation for capable, efficient on-device intelligence.