below marketmost similar roles pay hereabove market
This role pays more than
84%
of similar roles. Most pay
$202,800–$254,750
— the blue band above.
At the midpoint, this role pays about
$270k
versus about
$229k
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
3497 open roles
on FindRole.
Listed pay typically runs
$174,149–$277,600
across 2702 roles with salary data.
The AI Research Scientist: Multimodal Foundation Models - Architecture, Pre-Training & Distillation joins the Multimodal Intelligence Team to advance foundation models for Apple experiences. You will investigate fundamental questions regarding how multimodal models should be designed, trained, and distilled to balance broad intelligence with on-device constraints like memory, latency, and energy. Daily responsibilities include developing and evaluating new model architectures, pre-training objectives, data strategies, and teacher–student learning techniques. You will work with PyTorch or JAX to explore dense, recurrent, state-space, and mixture-of-experts architectures. The role focuses on co-developing frontier models and efficient models for Apple silicon, ensuring that capabilities like reasoning and multimodal understanding are preserved when transferring knowledge from large models to smaller, more efficient versions for private, integrated intelligent experiences.
Develop and evaluate new model architectures, pre-training objectives, and data strategies for multimodal foundation models.
Design teacher-student learning techniques to transfer capabilities from large models to smaller, efficient models.
Conduct large-scale pre-training experiments for models spanning language, images, video, audio, and sensor data.
Incorporate memory, latency, and energy constraints into the research process for on-device deployment on Apple silicon.
Implement architecture and algorithm co-design for memory-efficient and energy-efficient inference.
Execute offline and on-policy distillation using teacher-generated data, logits, and representation signals.
Evaluate pre-trained models across language understanding, reasoning, instruction following, and multimodal capabilities.
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.
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.
Strong 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.
Experience with knowledge distillation, teacher-student training pipelines, or transferring capabilities from large models to smaller models (preferred).
PyTorch
JAX
LLM Pre-training
Multimodal Foundation Models
Knowledge Distillation
Mixture-of-Experts
State-Space Models
Distributed Training
Model Architecture
Data Mixtures
Tokenization
Scaling Laws
hardware–software co-design
Apple Silicon
Python
PyTorch
LLM Pretraining
Distributed Computing
Spark
Ray
Deep Learning
Synthetic Data Generation
Data Curation
Machine Learning Systems
Data Processing Pipelines
Model Training
Data Filtering
Deduplication
Scaling Laws
Generative AI
Multimodal Foundation Models
Deep Learning
Python
PyTorch
JAX
Computer Vision
Data Pipelines
On-device Machine Learning
Data Curation
Vision-Language Models
Video-Language Models
Python
PyTorch
JAX
Deep Learning
Generative AI
Multimodal Foundation Models
Computer Vision
Data Pipelines
On-device Machine Learning
Data Curation
Vision-Language Models
Video-Language Models