Staff ML Engineer: Large-Scale Training & AI Platform

Apple, Inc.; Seattle (WA); On-site; USD 175,000 - 309,000; Full time

Job summary

Apple Cloud AI Platform seeks a Staff ML/Research Engineer to lead and shape the model training stack for large-scale projects on GPUs/TPUs. You will work with researchers and engineers to craft training recipes for pre-training, fine-tuning, RL, and distillation, applying parameter-efficient methods and stability tuning.

You will guide cross-team technical direction, mentor engineers, and contribute to high-performance, privacy-conscious training pipelines across Apple products.

Qualifications

• 8+ years of experience in large-scale model training, including reinforcement learning.
• 5+ years of experience building and operating large-scale model training systems in production.
• Hands-on experience with fine-tuning and distillation techniques such as supervised fine-tuning and LoRA.
• Proficiency with high-level training frameworks such as Megatron-LM, DeepSpeed, TorchTitan, NeMo, AXLearn, or MaxText.
• Strong Python or C++ programming skills and distributed training knowledge.
• Academic degree in CS/CE or equivalent practical experience.

Responsibilities

• Develop cutting-edge ML models with researchers and engineers and translate needs into training features.
• Implement model distillation and compression methods to transfer large-model quality to smaller models.
• Train models with long-context, mixture-of-experts, and multimodal inputs, addressing stability and efficiency.
• Diagnose training issues such as loss spikes and numerical instability and tune precision and optimizer settings.
• Build scalable training frameworks and services on GPUs/TPUs using PyTorch or JAX and higher-level frameworks.
• Create RL training infrastructure for post-training, including rollout, inference, and secure sandboxes.

Skills

Python

C++

PyTorch

JAX

Distributed training

Leadership

Communication

Education

BS/MS/PhD in CS or related

Tools

Megatron-LM

DeepSpeed

TorchTitan

NeMo

AXLearn

MaxText

Job description

Apple Cloud AI Platform seeks a Staff ML/Research Engineer to lead and shape the model training stack for large-scale projects on GPUs/TPUs. You will work with researchers and engineers to craft training recipes for pre-training, fine-tuning, RL, and distillation, applying parameter-efficient methods and stability tuning.

You will guide cross-team technical direction, mentor engineers, and contribute to high-performance, privacy-conscious training pipelines across Apple products.