Offensive Security Researcher – AI Systems

About the role

Apple’s Security Engineering & Architecture (SEAR) team is looking for an Offensive Security Researcher to focus on AI‑powered features and autonomous systems across Apple platforms. You will proactively discover and exploit vulnerabilities in large‑language models, prompt‑injection vectors, and AI‑driven workflows that could affect billions of users.

Key responsibilities

• Identify, develop and demonstrate exploits targeting AI/LLM components, agentic systems and related data‑exfiltration pathways.
• Conduct offensive research on emerging AI attack classes such as prompt injection, model manipulation and AI‑assisted privilege escalation.
• Build and refine tooling and pipelines that automate AI‑specific vulnerability discovery.
• Collaborate with world‑class security researchers to shape Apple’s AI security posture.
• Document findings, provide proof‑of‑concept code and recommend mitigations to product teams.

Required profile

• Strong foundation in common vulnerability classes (memory corruption, logic flaws, authentication bypass).
• Proven experience in security research or offensive security on complex platforms (browsers, messaging, distributed systems or AI platforms).
• Deep understanding of modern AI/LLM architectures and their failure modes.
• Experience applying AI/ML tools to augment security research workflows.

Required skills

• Programming: Python, C, C++, Swift, Objective‑C.
• Reverse‑engineering tools: IDA, Ghidra, LLDB.
• Familiarity with Apple operating systems (iOS, macOS) and their security models.
• Knowledge of AI/LLM prompt‑injection techniques and model manipulation strategies.