As a member of the Responsible AI and Safety group, you will be working on Apple's generative models that will power a wide array of new features, as well as longer term research in the generative AI space. Our team is currently interested in large generative models for vision and language, with particular interest on Responsible AI, safety, fairness, robustness, explainability, and uncertainty in models.

Apple Intelligence is powered by thoughtful data sampling, creation, and curation; high quality, detailed annotations; and application of these data to evaluate and mitigate safety concerns of new generative AI features. This role heavily draws on applied data science, scientific investigation and interpretation, cross-functional communication and collaboration, and metrics reporting and presentation to stakeholders & decision-makers. Responsibilities include:

Develop metrics for evaluation of safety and fairness risks inherent to generative models and Gen-AI features
Design datasets, identify data needs, and work on creative solutions, scaling and expanding data coverage through human and synthetic generation methods
Develop sampling strategies and combine human annotation with auto-grading to deliver high-quality, high-confidence insights at a fast pace and large scale
Use and implement data pipelines, and collaborate cross-functionally to execute end-to-end safety evaluations
Distill project findings into recommendations for product engineering teams and safety policy development
Develop ML-based enhancements to red teaming, model evaluation, and other processes to improve the quality of Apple Intelligence’s user-facing products
Work with highly-sensitive content with exposure to offensive and controversial content

Minimum Qualifications

MS or PhD in Computer Science, Linguistics, Cognitive Science, HCI, Psychology, Mathematics, Physics, or a similar science or technology field with a strong basis in scientific data collection and analysis + at least 4 years of relevant work experience, or BA/BS with 8+ years of relevant work experience
Experience gathering and analyzing language data, image data, and/or multi-modal data, including LLM-generated data
Strong experience designing human annotation projects, writing guidelines, and dealing with highly multi-labeled, nuanced, and often conflicting data
Proficiency in data science, machine learning, analytics, and programming with Python & Pandas; strong experience with one or more plotting & visualization libraries
Ability to collaborate with team members to prioritize competing projects, set and maintain a schedule for milestones and project completions, and communicate with all levels of team members as well as external stakeholders
Strong skills for rigorous model quality metrics development; interpretation of experiments and evaluations; and presentation to executives

Preferred Qualifications

Experience working in the Responsible AI space
Curiosity about fairness and bias in generative AI systems, and a strong desire to help make the technology more equitable
Prior scientific research and publication experience
Experience working with generative models for evaluation and/or product development, and up-to-date knowledge of common challenges and failures
Proven track record of contributing to diverse teams in a collaborative environment
A passion for building outstanding and innovative products. This position involves a wide variety of interdisciplinary skills