RemoteLeaf Logo
Log in Sign up

AI Agent Security Research Engineer

2 months ago
Asia, Singapore, Hong Kong
Full-time
Senior
AI (Artificial Intelligence)
Artificial Intelligence and Machine Learning
DevSecOps Docker GitHub Go Java Jenkins Kubernetes Microservices Python SonarQube Tekton
Apply Now
OKX

OKX

OKX operates as a leading cryptocurrency exchange, providing users with a platform to buy, sell, and trade various digital assets such as Bitcoin, Ethereum, and XRP, while also offering tools for exploring Web3, decentralized finance (DeFi), and non-fu...

Diversified Financial Services
1K-5K
Founded 2017
View All Jobs 81

Description

  • Design and implement a multi-agent code auditing system for vulnerability detection, malicious code identification, and sensitive information leakage.
  • Define Planner, Executor, and Critic roles, tool invocation chains, and cross-agent state synchronization mechanisms.
  • Integrate RAG, Chain-of-Thought, Reflection, and related techniques into security audit agents to improve detection performance.
  • Build quantifiable evaluation and iteration frameworks to continuously improve accuracy and recall.
  • Develop plugins for GitLab CI/CD, Tekton, and Jenkins to enable audit-on-commit within DevSecOps workflows.
  • Construct a security protection framework for LLM applications across input, output, and runtime layers.
  • Develop agent workflows for automated alert classification, contextual correlation, and false positive filtering.
  • Integrate threat intelligence retrieval and automated analysis outputs to support SOAR platform integration.
  • Design human-machine collaboration intervention mechanisms and agent behavior audit systems for production environments.
  • Build highly available, scalable agent service architecture for concurrent scanning, fault tolerance, API outputs, rule management, result visualization, and false positive feedback.

Requirements

  • 3+ years of backend development experience.
  • Proficiency in at least one of Python, Go, or Java.
  • Hands-on experience deploying LLM agents in production, including agent architecture design, hallucination handling, and tool invocation fault tolerance.
  • Hands-on experience with AI security, including prompt injection, jailbreaking, malicious agent injection, and tool misuse defenses.
  • Familiarity with at least one agent framework such as LangChain, LlamaIndex, AutoGen, CrewAI, or LangGraph, with production project experience.
  • Proficient in Docker and Kubernetes.
  • Experience with microservices architecture design and deployment.
  • Experience with SAST/SCA tools or code auditing tools such as CodeQL, Semgrep, or SonarQube (preferred).
  • Experience with LLM fine-tuning such as SFT or LoRA, or local deployment and optimization of models like Llama 3, Qwen, or DeepSeek (preferred).
  • Open-source contributions related to agents or pull requests to mainstream LLM frameworks, security competition awards, or CVE/CNVD submissions are preferred.

Benefits

  • Competitive total compensation package.
  • L&D programs and education subsidy for employee growth and development.
  • Various team-building programs and company events.
  • Wellness and meal allowances.
  • Comprehensive healthcare schemes for employees and dependants.

Interested in this position?

Apply directly on the company website

Apply Now

Similar Roles

Mindrift.ai: Be the “I” in AI

Mortgage Underwriter - Freelance AI Trainer

Mindrift.ai: Be the “I” in AI Internet Software & Services

Mindrift is seeking mortgage underwriting and loan origination professionals for project-based AI evaluation work focused on testing and improving mortgage-related AI outputs and compliance decisions.

Greece Portugal Italy Part-time Mid Level AI (Artificial Intelligence) Financial Analyst
Up to $104k
15 hours, 53 minutes ago
Keywords Studios

Active Directory / Identity Engineer

Keywords Studios 10K-50K Internet Software & Services

Keywords Group is seeking an experienced Active Directory SME and Azure Identity Lead to guide company-wide identity architecture and support global IT across on-premises and cloud environments.

United Kingdom Full-time Lead Security Engineer
Active Directory Cybersecurity DHCP DNS PowerShell
15 hours, 53 minutes ago
Mindrift.ai: Be the “I” in AI

Claims Processing Agent - Freelance AI Trainer

Mindrift.ai: Be the “I” in AI Internet Software & Services

Mindrift is seeking part-time project-based insurance and claims specialists to evaluate and improve AI systems for auto insurance decision-making, fraud detection, and subrogation testing.

United States Part-time Mid Level AI (Artificial Intelligence)
Up to $125k
15 hours, 53 minutes ago
Toloka

Record Your Daily Routine & Get Paid - AI Training (Remote)

Toloka 251-1K Internet Software & Services

Project-based freelance opportunity with an AI training platform recording first-person videos of everyday household activities to help train AI systems and robots.

Guatemala Freelance Entry Level AI (Artificial Intelligence)
$12k-$12k
15 hours, 53 minutes ago

You're on a roll! Sign up now to keep applying.

Sign Up

Already have an account? Log in

Used by 14,729+ remote workers