AIML - Machine Learning Engineer, Red Teaming - Responsible AI and Safety

Would you like to play a part in protecting Apple Intelligence from real-world misuse before it reaches users? Were looking for a Machine Learning Engineer to join Apples Responsible AI red-teaming function - a dedicated team focused on actively probing Apple Intelligence features to find real safety vulnerabilities. In this role youll own a cluster of AI features, invent novel attacks against them, and build the tooling to scale your own techniques. This role focuses on actively red-teaming Apple Intelligence features - pre-ship and post-ship - to find real, reproducible vulnerabilities. Youll take in signals from feature risk assessments, policy definitions, evaluation results, code changes, and published attack research, and use them to probe for novel failure modes across features and models, including attacks in agentic scenarios.Producing red-teaming findings that meaningfully improve Apples Responsible AI posture requires validating that an issue is real and reproducible, triaging its severity and reach, and driving it to resolution through the right stakeholders. This role also involves building automated red-teaming tooling - attack generation, batch orchestration, and feedback loops against known issues - to scale beyond manual testing, and partnering with a dedicated crowd and with adjacent security teams to scale efforts as well as share findings and best practices.This role heavily draws on adversarial and creative technical thinking, applied security and linguistic expertise, cross-functional communication, and sustained ownership of a feature area over time. Prior experience in AI red-teaming recommended.Own a dedicated cluster of Apple Intelligence features and build sustained expertise in their architecture and failure modesActively probe systems pre-ship and post-ship to find real, reproducible vulnerabilitiesInvent novel attack strategies and keep pace with newly published jailbreak and attack researchValidate and triage findings by severity and reach, and drive them through to mitigationDesign and build automated red-teaming tooling to scale attack generation and testing coverageIdentify gaps in policy and evaluation coverage and feed them back to eval and mitigation teamsPartner with the internal and external crowd and and with adjacent security teams to route findings appropriatelyWork with highly sensitive content, including exposure to offensive and controversial material.MS, or PhD in Computer Science, Machine Learning, Statistics, or related fields; or an equivalent qualification acquired through other avenues.Experience working with generative models for evaluation and/or product development, and up-to-date knowledge of common challenges and failures.Strong engineering skills and experience in writing production-quality code in Python.Deep experience in foundation model-based AI programming (i.e.: using DSPy for optimizing foundation model prompts, for example) and a drive to innovate in this space.Experience working with noisy, crowd-based data labels and human evaluations.Experience working in the Responsible AI space.Prior scientific research and publication experience.Strong organizational and operational skills working with large, multi-functional, and diverse teams.Curiosity about fairness and bias in generative AI systems, and a strong desire to help make the technology more equitable.