The Operational Security Automation role evolves in 2024 to integrate generative AI and agentic AI as core drivers of security center operations. This position transforms traditional SOCs or VOCs into autonomous operational centers capable of contextual reasoning, decision-making, and action.
Key Responsibilities :
1. Intelligent AI Workflow Development
- Design of self-adaptive playbooks using LLMs (GPT-4, Claude, Mistral)
- Creation of orchestrated APIs for autonomous agentic workflows
- Integration of MCP (Model Context Protocol) and Agent2Agent protocols
- Development of AI agents for contextual automatic incident triage
2. AI Autonomous Operations Governance
Supervision of autonomous decisions with human validation mechanismsROI measurement of deployed generative AI systemsCompliance with AI Act, DORA, and NIS2 frameworks for autonomous AIPerformance management according to agentic SLAs / SLOs3. AI Strategy and Innovation
Development of strategic roadmap for agentic AI implementationTechnology watch on generative model evolutionIntegration of innovative perspectives from AI threat landscapeBenchmarking of SOAR platforms with agentic capabilities4. AI Performance Management
Definition of specific KPIs for generative systemsAnalysis of contextual relevance of autonomous decisionsMeasurement of automatically generated playbook effectivenessContinuous model optimization through fine-tuning5. AI Skills Development
Planning of required competencies for the agentic AI eraContinuous training on fine-tuning and LLM optimizationManagement of specialized technical resources in generative AICreation of AI upskilling programsRequired Experience :
10-12+ years in information Security with cloud and AI focus5+ years of experience in managing a team of SOAR or SIEM members.Mastery of agile methodologies adapted to AI cyclesExperience in Agentic SOAR Platforms
Tines AI with generative capabilitiesXSOAR with Cortex XSIAM and integrated AIIBM Resilient with advanced Watson AISwimlane with agentic modulesAI Protocols and Standards
Model Context Protocol (MCP) - AnthropicAgent2Agent (A2A) - GoogleAI PERFORMANCE INDICATORS :
Generative Metrics
Automatic playbook generation rateGenerated decision quality (precision / recall)Response time reduction through AIMeasurable ROI of AI investmentsAgentic Metrics
Validated autonomous decision rateContainment latency with AI agentsIncidents resolved without human interventionPerformance of self-adaptive systemsOperational Metrics
SLA / SLO compliance with AI systemsAutomatic threat pattern coverageScalability of deployed agentic solutionsTeam adoption rates of AI toolsAI REGULATORY CONTEXT
Required Compliance
EU AI Act regulationDORA directives for digital financeNIS2 for network securityAI Risk Governance
Mapping of specific agentic AI risksProcedures for validating autonomous decisionsAI audit and traceability mechanismsContinuity plans for AI failures