AIML - Machine Learning Engineer, Red Teaming - Responsible AI and Safety
Core
Red-team Apple Intelligence features to find and mitigate real-world safety vulnerabilities and misuse before they reach users.
Role type
Senior IC machine learning engineer (AI safety/red-teaming)
Builds
Automated red-teaming tooling for attack generation, batch orchestration, and feedback loops
Domain
Generative AI, Responsible AI, Adversarial Security
Deliverable
production ML models
Required skills
Generative model evaluation, Python production code, Foundation model programming (e.g., DSPy), Adversarial attack research, Crowd-based data labeling, Human evaluation design
Preferred skills
Responsible AI domain experience, Scientific research publications, Organizational skills for large teams, Fairness and bias analysis
Technologies
Python, DSPy, Foundation Models
Responsibilities
Own a cluster of AI features and build expertise in their architecture and failure modes; Probe systems pre-ship and post-ship to find reproducible vulnerabilities; Invent novel attack strategies and track new jailbreak research; Validate and triage findings by severity and reach; Design automated tooling to scale attack generation; Identify gaps in policy and evaluation coverage; Partner with security teams and external crowds to route findings.
Seniority
Senior, hands-on IC