This role pays more than
76%
of similar roles. Most pay
$208,833–$270,500
— the shaded band above.
At the midpoint, this role pays about
$270k
versus about
$240k
for comparable roles.
Based on 240 similar postings.
Employer
About Apple Inc
Apple Inc. is a multinational technology company known for designing and manufacturing consumer electronics, software, and online services, including the iPhone, Mac, iPad, and App Store. Industry: Consumer Electronics & Software
Apple Inc currently has
1984 open roles
on FindRole.
Listed pay typically runs
$175,000–$277,600
across 1590 roles with salary data.
As part of the Multimodal Intelligence Team, the AI Research Scientist: Multimodal Foundation Models - Architecture, Pre-Training & Distillation will advance architectures, pre-training methods, and distillation techniques for multimodal models. The role involves investigating fundamental questions regarding model design, training objectives, data strategies, and teacher–student learning to transfer capabilities from large foundation models to smaller, efficient versions. You will develop and evaluate new architectures, including dense, recurrent, state-space, and mixture-of-experts models, while focusing on inference efficiency for Apple silicon. Key responsibilities include managing the full lifecycle of model development, from data mixtures and tokenization to scaling and evaluation across language, image, video, audio, and sensor-derived representations. Required skills include proficiency in PyTorch or JAX, distributed training systems, and experience with transformer-based architectures to solve challenges related to memory, latency, and energy constraints for on-device deployment.
Design and evaluate new multimodal foundation model architectures, including dense, recurrent, state-space, and mixture-of-experts models.
Develop pre-training objectives, data mixtures, and training curricula for large-scale multimodal models.
Implement teacher-student learning techniques to transfer capabilities from large models to smaller, efficient versions.
Optimize model architectures specifically for memory, latency, and energy constraints on Apple silicon.
Conduct large-scale experiments to improve reasoning, instruction following, and multimodal understanding in compact models.
Translate research findings into production-ready technologies for integration across the Apple ecosystem.
Publish research findings and open-source selected models, tools, and evaluation artifacts.
What we're looking for
Master's degree or equivalent practical experience in machine learning, computer science, or a related technical field.
Minimum of 6 years of relevant experience in the field.
Hands-on experience designing, implementing, and running large-scale pre-training experiments for large language models.
Experience with LLM pre-training topics including architecture, training objectives, data mixtures, tokenization, scaling, and optimization.
Proficiency with modern deep learning frameworks such as PyTorch or JAX and distributed training systems.
Experience evaluating pre-trained models across language understanding, reasoning, instruction following, or multimodal capabilities.
Strong understanding of transformer-based architectures and current approaches to efficient or scalable foundation-model training.
Preferred experience in knowledge distillation, multimodal models, inference efficiency, and research publications or influential open-source contributions.
Python
PyTorch
JAX
TensorFlow
Deep Learning
Foundation Models
Large Language Models
Reinforcement Learning
Vision-Language Modeling
Video Generation
Data Pipelines
Reward Modeling
Multimodal Models
On-policy Distillation
Python
PyTorch
JAX
TensorFlow
Deep Learning
Foundation Models
Large Language Models
Reinforcement Learning
Vision-Language Modeling
Video Generation
Data Pipelines
Reward Modeling
On-policy Distillation
Multimodal Models