AI Security Engineer

Net2Source (N2S) is seeking an AI Security Engineer responsible for designing, securing, and deploying agentic AI systems within enterprise environments. This role involves a blend of security engineering, threat modeling, and applied AI system development to ensure intelligent agents operate reliably and securely.

Responsibilities:

• Architect security controls (identity, network, runtime isolation, sandboxing, policy enforcement) for agent-based AI systems
• Develop AI specific threat models addressing agent misbehaviour, adversarial prompts, data leakage, model tampering, and supply chain risks
• Evaluate third party AI tools, APIs, and agent frameworks for security compliance and risk
• Design and implement autonomous AI agents using LLMs, APIs, orchestration frameworks, and multi agent systems
• Build agent behaviour logic including tool use, routing, planning, fallbacks, and guardrails
• Prototype and iterate AI agents in production, refining reliability, safety, and output quality based on real world usage
• Develop security monitoring pipelines for agent executions and automate detection of anomalous or harmful agent behaviour
• Implement secure MLOps practices—including model lineage tracking, training data protection, and integrity controls
• Perform vulnerability assessments, penetration testing, and red teaming of AI agents and underlying infrastructure
• Work closely with AI research, product, engineering, cloud, and cybersecurity teams to ensure agents are performant, safe, and compliant
• Translate business workflows into agent behaviours through scoping, discovery sessions, and requirements definition
• Establish secure development standards for agentic AI systems and contribute to enterprise AI governance frameworks
• Publish internal best practices for agent security, including prompt security guidelines, LLM threat mitigation, and safe tooling patterns

Requirements:

• 8+ years in cybersecurity engineering, application security, or cloud security
• Hands on experience with LLMs, AI/ML pipelines, vector databases, orchestration frameworks (AutoGen, CrewAI, LangGraph, etc.)
• Strong programming background (Python required; Java/C++ optional)
• Expertise in threat modelling, identity & access management, secure API design, and network segmentation
• Familiarity with adversarial ML, model robustness testing, data poisoning defences, and model evaluation
• Experience deploying secure workloads in AWS/Azure/GCP
• Experience building autonomous agents or multi agent systems
• Knowledge of AI governance, safety, and responsible AI frameworks
• Background in cryptography, secure CI/CD pipelines, MLOps, and privacy preserving ML