AI Agent Security Researcher

About the company:

We have partnered with a renowned global leader in information and communications technology (ICT) infrastructure and smart devices. They are providing full-stack, all-scenario solution for products and services carriers, enterprises, governments, and individual consumers worldwide.

Our client is looking for an AI Agent Security Researcher to join the team.

Job Overview

As HarmonyOS evolves into an intelligent operating system, it increasingly integrates AI

technologies to enhance user and industry services. However, AI-powered features—such as AI

Agents and large language models (LLMs)—introduce new security risks, including prompt

injection attacks, unauthorized API access, and data leakage. Since AI Agents often operate with

high-privilege permissions, improper access control or oversharing could lead to system

compromise, privacy breaches, or unexpected behaviors.

To address these challenges, we are expanding our AIOS Security R&D team to strengthen system-level protections for HarmonyOS. This role focuses on designing and implementing security

mechanisms to Safeguard AI Agent. Candidates will work on AI agent security and Access Control

frameworks, especially Dynamic Least Privilege, AI Fence, Intend-based Sandbox, MCP and Skill

security on so on, to prevent exploitation in mobile, PC, and IoT environments.

1. Job Requirements:

Either of following Technical Expertise are welcome,

Proficiency in C/C++ for system programming (e.g., kernel modules).

Deep understanding of OS security mechanisms, including:

• Mandatory Access Control (MAC) (e.g., SELinux, AppArmor)
• Kernel hardening (memory protection, syscall filtering)
• Secure API gateways for AI Agent interactions

Experience with AI/LLM security risks, such as:

• Prompt injection attacks
• Adversarial ML exploitation
• Over-privileged AI Agent behaviors

Familiarity with mobile/PC security frameworks (Android SE, iOS Sandbox, Linux

security modules).

2. Qualifications

• Master's or PhD in Computer Science, Cybersecurity, or AI/AI agent security.
• Industry experience in OS security (e.g., Android/iOS/Linux kernel security).
• Publications or patents in cybersecurity (preferred but not mandatory).

3. Preferred Experience

• Worked on AI Agent.
• Contributed to system security projects (e.g., Linux kernel, Android AOSP).

Interested candidates, who wish to apply for the advertised position, please click on Apply Now. We regret that only shortlisted candidates will be notified.

EA License No.: 01C4394

By sending us your personal data and curriculum vitae (CV), you are deemed to consent to PERSOL Singapore Pte Ltd and its affiliates to collect, use and disclose your personal data for the purposes set out in the Privacy Policy available at https://www.persolsingapore.com/policies. You acknowledge that you have read, understood, and agree with the Privacy Policy