Video-Gen ML Engineer: Scale Large Generative Models

Apple Inc.; San Diego (CA); On-site; USD 142,000 - 263,000; Full time

Job summary

Apple Inc. in San Diego is seeking a machine learning engineer to advance video generation models from pre-training to efficient inference. You will design training recipes, manage large distributed runs, and push for reproducible, production-ready code across Apple teams.

The role emphasizes hands-on experience with large generative models for video, distributed GPU clusters, and proficiency in Python with PyTorch or JAX. Collaboration with researchers and product teams is key.

Qualifications

• Bachelor's degree in EE/CS/CE or related field with 3+ years of industry experience.
• Experience training large generative models for video generation.
• Experience running distributed training on multi-node GPU clusters.
• Proficiency in Python with PyTorch or JAX.

Responsibilities

• Pre-training video generation models, including architecture decisions and training recipe design.
• Run, monitor, and debug large-scale distributed training jobs and diagnose issues.
• Improve training efficiency with parallelism, mixed-precision, checkpointing and hardware utilization.
• Improve inference efficiency through distillation, sampling, and quantization.
• Adapt pre-trained models via fine-tuning and knowledge distillation.
• Build and maintain training and evaluation code with emphasis on reproducibility.
• Partner with product and research teams to translate requirements into modeling tasks.

Skills

Python programming

Distributed training

Large-scale training experience

Education

Bachelor's degree in Electrical Engineering, Computer Science, Computer Engineering, or relevant degree

Tools

PyTorch

JAX

Job description

Apple Inc. in San Diego is seeking a machine learning engineer to advance video generation models from pre-training to efficient inference. You will design training recipes, manage large distributed runs, and push for reproducible, production-ready code across Apple teams.

The role emphasizes hands-on experience with large generative models for video, distributed GPU clusters, and proficiency in Python with PyTorch or JAX. Collaboration with researchers and product teams is key.