Description
We are looking for people with a track record in building models and model-driven products to affect user experiences. Join us, and impact hundreds of millions of customers across billions of their interactions with foundation model powered Apple Intelligence features, that are available on iPhone, iPad, HomePod, Mac, Watch, CarPlay, and tv across more than 30 languages.
- Algorithm development: Define signals that are important in prompts, responses and CoT reasoning steps. These usually require a fine-tuned model for specific use cases.
- Model evaluation: Understand the importance of a balanced eval-set. Ability to perform error analysis to figure out how to improve model capabilities
- Ablation experiments: Test your data augmentation strategies via ablation experiments. Comfortable debugging training errors, and tune hyper-parameters and data mixture to achieve desired outcome.
- Data processing and data filtering: Ability to efficiently process and filter very large amounts of data, often times messy.