AI Security Engineer

About the role

Systems Limited is seeking an experienced AI Security Engineer to protect AI‑powered applications, large language models, AI agents, and retrieval‑augmented generation pipelines. You will work on identifying and mitigating AI‑specific threats such as prompt injection, jailbreaks, data leakage, and unsafe model outputs.

Key responsibilities

• Design and implement security controls for AI applications, LLM integrations, AI agents, and RAG‑based systems.
• Identify and mitigate AI‑specific threats including prompt injection, jailbreaks, data exfiltration, and hallucination risks.
• Secure prompt workflows, system prompts, agent instructions, retrieval pipelines, and AI orchestration layers.
• Assess risks in LLM APIs, vector databases, embeddings, knowledge bases, and external tool integrations.
• Conduct threat modeling, red‑team exercises, and security audits for AI solutions.
• Define and enforce AI security governance, responsible AI controls, and compliance best practices.
• Monitor emerging AI‑related security threats and recommend proactive mitigation strategies.
• Collaborate with engineering teams to secure AI APIs, cloud deployments, access controls, and production AI platforms.

Required profile

• 6‑9 years of experience securing AI applications, LLMs, RAG pipelines, and AI agents.
• Hands‑on understanding of AI security risks such as prompt injection, jailbreaks, and data leakage.
• Strong knowledge of secure AI architecture, including authentication, authorization, logging, monitoring, and data protection.
• Experience with cloud platforms (Azure, AWS, or GCP) and AI service deployments.
• Familiarity with security tooling and frameworks for AI systems.

Required skills

• Vector search
• Embeddings
• LLM red teaming
• OWASP Top 10 for LLM Applications
• Prompt engineering
• Responsible AI governance
• AI risk management
• AI guardrails
• Data privacy compliance
• AI security frameworks
• SAST
• Prompt filtering
• Security monitoring
• RAG
• CISSP
• AI agents
• Content safety
• Output validation
• DAST
• Vulnerability scanning
• SIEM