AI Agent Security Engineer; OS Security​/Access Control​/Trust Framework

AI Agent Security Engineer (OS Security / Access Control / Trust Framework)

We are building a next-generation intelligent computing platform that integrates advanced AI capabilities to power seamless user and enterprise experiences. As AI agents and large language models (LLMs) become core components of modern systems, they also introduce new and complex security challenges—such as prompt injection, unauthorized API access, and data leakage.

Given that AI agents often operate with elevated privileges, improper access control or excessive data exposure can lead to system compromise, privacy risks, and unintended behaviors.

To address these challenges, we are expanding our AI Security R&D team to design and implement robust, system-level protections for AI-driven environments across mobile, desktop, and IoT platforms.

• Design and implement security mechanisms to safeguard AI agents and LLM-powered systems
• Develop and enhance access control frameworks
  , including dynamic least-privilege models and sandboxing mechanisms
• Secure AI system interactions
  , including APIs, plugins, and tool integrations
• Identify, analyze, and mitigate emerging threats in AI systems (e.g., prompt injection, adversarial attacks)
• Collaborate with cross-functional teams to integrate security into system architecture and AI workflows

• Proficiency in C/C++ for low-level or system programming (e.g., kernel modules, system services)

• Deep understanding of OS security mechanisms
  , including:
  Mandatory Access Control (e.g., SELinux, App Armor), kernel hardening (memory protection, syscall filtering), secure API design and enforcement

• Familiarity with AI-specific security risks
  , including: prompt injection attacks, adversarial machine learning techniques, risks from over-privileged AI agents

• Experience with security frameworks across mobile, desktop, or Linux-based systems

• Master's or PhD in Computer Science, Cybersecurity, Artificial Intelligence, or related fields
• Industry experience in operating system or platform security (e.g., Linux, Android, iOS)
• Strong problem-solving skills and a security-first mindset
• Publications or patents in cybersecurity are a plus (not required)

• Hands‑on experience working with AI agents or AI-enabled systems
• Contributions to system-level or open-source security projects (e.g., Linux kernel, Android AOSP)
• Experience designing or implementing access control or sandboxing frameworks