Generative AI RL Research Engineer (LLMs & Tooling)

Apple Inc.; Zürich; Vor Ort; CHF 120.000 - 180.000; Vollzeit

Zusammenfassung

Apple Inc. in Zurich invites a Machine Learning Research Engineer to join the Generative AI team specializing in foundation model post-training.

You will advance RL for agentic tool use, planning and reasoning to strengthen Apple’s foundation models and influence features across Siri. The role involves scaling RL methods, training agents with tool use and API integration, building reward models and simulators, running large-scale experiments, and collaborating with Europe-based RL/ML experts and

Qualifikationen

• MSc/PhD or equivalent research/industry experience in Computer Science, ML, EE, or related field.
• Hands-on reinforcement learning and deep learning with large-scale models, especially LLMs.
• Proficiency in Python and ML frameworks (PyTorch, JAX), with distributed training experience.

Aufgaben

• Develop RL methods to improve reasoning, instruction following, multi-turn dialogue, and reduce hallucinations in large language models.
• Design and train agents with tool use, planning, and API integration to reliably accomplish tasks.
• Build and refine reward models, evaluators, datasets, and simulation environments (e.g., RLHF, RLAIF, RLVF).
• Run large-scale experiments, analyze results, and translate findings into research contributions and practical improvements for Apple Intelligence.
• Collaborate within a Europe-based team of ~35 RL/ML experts, coordinating with US foundation model groups.

Kenntnisse

Reinforcement learning

Deep learning

Python

Distributed training

Communication

Ausbildung

MSc/PhD in CS/ML/EE

Tools

PyTorch

JAX

Jobbeschreibung

Apple Inc. in Zurich invites a Machine Learning Research Engineer to join the Generative AI team specializing in foundation model post-training.

You will advance RL for agentic tool use, planning and reasoning to strengthen Apple’s foundation models and influence features across Siri. The role involves scaling RL methods, training agents with tool use and API integration, building reward models and simulators, running large-scale experiments, and collaborating with Europe-based RL/ML experts and