Role overview
As a Machine Learning Engineer for LLM Agent and Reinforcement Learning, you will play a pivotal role in shaping the next generation of intelligent agent systems that power Apple Intelligence. You will develop advanced LLM-based agents capable of reasoning, planning, and acting across multi-turn tasks and tool-use environments. Your work will directly influence Apple’s foundation models and agentic intelligence capabilities across products and platforms.
You will collaborate with cross-functional research and engineering teams to push the frontier of agent modeling, mid-training, reinforcement learning, and system-level orchestration. Your responsibilities will include designing innovative mid-training and post-training pipelines, scaling large-scale synthetic data generation for agent learning, and improving end-to-end agent behaviors through data-model co-design. You will also have the opportunity to publish and present your work at top ML and AI research venues.
Your work may span a variety of directions, including but not limited to:
Develop and improve LLM-based agent models that can plan, reason, and act across complex, multi-turn tasks.
Design mid-training pipelines to enhance agentic capabilities such as tool-use, reflection, and long-horizon reasoning.
Explore and implement reinforcement learning and feedback-based fine-tuning to align agent behaviors with human and system objectives.
Build simulation and evaluation environments for measuring multi-turn success rate, planning efficiency, and robustness of agent interactions.
Develop scalable frameworks for synthetic agentic data generation, including user-assistant-tool trajectories, environment modeling, and self-play.
Collaborate with modeling, data, and infrastructure teams to integrate new training signals and improve overall system performance.
Stay on top of cutting-edge research in agentic LLMs, reasoning, tool-use, mid-training, and reinforcement learning.
What we're looking for
Strong background in machine learning, natural language processing, or reinforcement learning with a focus on LLM or agentic systems.
Solid understanding of LLM training paradigms (pre-training, mid-training, RL, post-training) and multi-modal or tool-use-enhanced architectures.
Excellent programming skills and experience with Python and modern deep learning frameworks (PyTorch, JAX, or equivalent).
Ability to design, run, and analyze large-scale training and evaluation experiments.
M.S. in Computer Science, Machine Learning, Artificial Intelligence, or related fields.
Apple is an equal opportunity employer that is committed to inclusion and diversity. We seek to promote equal opportunity for all applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, Veteran status, or other legally protected characteristics. Learn more about your EEO rights as an applicant .