AI Agent Security Research Engineer
Core
Design and implement AI-driven code security detection engines and LLM application security frameworks to protect against vulnerabilities, malicious code, and prompt injection attacks.
Role type
Senior IC AI Agent Security Research Engineer
Builds
Multi-agent collaborative code auditing systems, LLM security protection frameworks, and automated threat response workflows for DevSecOps pipelines.
Domain
Cybersecurity / AI Safety / Blockchain Infrastructure
Deliverable
production ML models | infrastructure
Required skills
Backend development (Python/Go/Java), LLM Agent architecture design, AI security (prompt injection, jailbreak defense), RAG and Chain-of-Thought integration, Microservices architecture, Docker and Kubernetes
Preferred skills
SAST/SCA tool experience (CodeQL, Semgrep), LLM fine-tuning (SFT, LoRA), Open-source agent framework contributions, CTF competition awards
Technologies
LangChain, LlamaIndex, AutoGen, CrewAI, LangGraph, GitLab CI/CD, Tekton, Jenkins, Docker, Kubernetes, RAG, SOAR
Responsibilities
Design multi-agent collaborative code auditing systems with role decomposition and state synchronization; Integrate RAG and reflection into security audit agents; Develop plugins for DevSecOps pipelines to achieve audit-on-commit; Construct security protection frameworks for LLM applications covering input/output/runtime layers; Develop Agent workflows for automated alert classification and false positive filtering; Design human-machine collaboration intervention mechanisms for Agent behavior audit.
Seniority
Senior, hands-on IC
