The Special Projects team at Apple is developing novel user-facing conversational features that leverage the multimodal capabilities of state-of-the-art foundation models. A key component of this process is the ability to produce complex simulated scenario data, in order to train and evaluate agentic AI models. We are looking for a skilled Data Scientist to work closely with our Simulation and Machine Learning Evaluations teams to generate large synthetic datasets, analyze the gap between simulated and real data, and evaluate and fine-tune agentic AI model performance at various tasks. A successful candidate is experienced in managing large, multi-modal datasets, in translating subjective product requirements into objective criteria, and has strong statistical analysis skills.

Work closely with ML Engineers to understand model evaluation needs

Work closely with the Simulation team to design and generate large, multi-modal datasets for model evaluation

Analyze the gap between simulated and real data

Collaborate with the Simulation team to prioritize and address simulation performance gaps