Job Details

Role Number: 200607600-2459

Summary

We build frontier foundation models that power intelligent experiences at Apple. Our team works across the full training lifecycle: including pre-training foundation models, and developing mid-training approaches that bridge general capability and task-specific performance. What makes our work distinct is that we're engineering models specifically for Apple silicon and optimized for experiences that are private, personal, and deeply integrated into the OS. We're solving frontier problems in reward modeling to resist reward hacking, handling sparse and delayed rewards in agentic settings, and aligning models reliably across the spectrum from open-ended creative tasks to precise, action-taking workflows. If you're drawn to hard problems where the research and the product are inseparable, this is the team

Description

We believe that the most interesting problems in deep learning research arise when we try to apply learning to real-world use cases, and this is also where the most important breakthroughs come from. You will work with a close-knit and fast growing team of world-class engineers and scientists to tackle some of the most challenging problems in foundation models and deep learning, including end-to-end voice interactions, natural language processing, and combining learning with knowledge.

Minimum Qualifications

• Demonstrated expertise in deep learning with publication record in relevant conferences (e.g., NeurIPS, ICML, ICLR, COLM, ACL, NAACL, EMNLP, CVPR, ICCV, ECCV, KDD, ACL, ICASSP, InterSpeech) or a track record in applying deep learning techniques to products

• Proficient programming skills in Python and one of the deep learning toolkits such as JAX, PyTorch, or Tensorflow

• Ability to work in a collaborative environment.

• PhD, or equivalent practical experience, in Computer Science, or related technical field.

Preferred Qualifications

• Code Large language models.

• Experience in building end-to-end voice models.

• Experience in speech recognitions and generations.

• Experience in reinforcement learning, on-policy distillation.

• Experience in post-training, mid-training large language models.