Agentic Security Engineer

Summary

As the first Agentic Security Engineer on Beazley Security’s MDR Operations Team, you’ll help us build a new managed service: providing our clients with visibility, control, and response for the agents that run and execute on their endpoints and servers.  You’ll work hands on with our agentic runtime control platform, helping us deploy it, continuously tune it, and ensure our analyst team is receiving actionable alerts that require immediate action and response. 

Reporting directly to the Director of MDR Operations, this role blends platform engineering, policy design, detection engineering, and analyst enablement.

You’ll work with MDR analysts, Product, Software Engineering, Professional Services, and vendor partners to shape how the service is configured, how its findings are routed and reported, and how it expands its coverage as AI adoption and capability keeps moving.

Ideal candidates bring experience from MDR/MSSPs, Software Engineering, or security product companies, along with hands on experience with Agentic applications such as Claude Code, Codex, OpenCode, OpenClaw, or Hermes. You know how hooks, skills, plugins, and MCP servers work because you’ve used them, and you’ve considered the risks that they introduce. 

Responsibilities

Platform Deployment & Management

• Deploy, configure, and maintain the AI security platforms that power our Managed AI Detection and Response solution, including rolling out the tooling through MDM or EDR and configuring the initial enforcement policies. 
• Manage the endpoint enforcement actions we run through the platform agent, including software removal and removal of MCP server, shell, and database integrations, with per-client pre-authorization and exclusion lists in place so enforcement never takes out business-critical tooling.
• Monitor platform health, coverage, webhook delivery, and enforcement-action lifecycle state so gaps in visibility or delivery are caught before they become gaps in detection or missed reports.

AI Policy & Configuration Engineering

• Translate client AI acceptable use policies, data handling requirements, and risk appetite into enforceable platform configurations under an “allow, ask, and deny” enforcement model.
• Build and maintain standard enforcement baselines by client segment, including exclusion lists and pre-authorized response actions we can take on clients’ behalf. 
• Tune client deployments and environment suppressions as usage patterns and feature adoptions evolve.  
• Keep policy changes documented, reviewed, and easy to understand. 

Detection, Response, and Automation. 

• Help build and tune the detection and routing logic that sends real threats to analysts for investigation and everything else to scheduled client reporting. 
• Use Python, platform APIs, and agentic tools like Claude Code to automate client onboarding, baseline policy deployment, enrichment, reporting, and common response actions so the service scales without adding labor.

Analyst Mentoring & Enablement

• Serve as the MDR team’s subject matter expert on AI risk, AI tooling, and the platform stack behind AIDR.
• Run and maintain analyst runbooks, promotion-trigger guidance, and training materials for analyst queue handling.
• Review analyst escalations with your peers to help identify areas of improvement or where additional guidance and training may be needed. 

Technology Evaluation & Service Development

• Track the AI threat landscape using the frameworks the service maps to, including MITRE ATT&CK, MITRE ATLAS, the OWASP Agentic Top 10, and the OWASP Top 10 for LLM Applications, and turn what you learn into service capabilities.
• Partner with MDR leadership and Product Engineering teams on the AIDR roadmap, and helping design the service as it grows. 

Required Qualifications

• 3+ years of experience in cybersecurity, including at least 1 year in an engineering detection engineering, or automation role, preferably within an MDR, MSSP, or security product company.
• Hands-on experience with generative AI tools, AI agents, and extensions such as hooks, skills, plugins, and MCP servers, along with a deep understanding of associated risks, including misalignment, unexpected behavior, prompt injection, credential exposure, and malicious skills or plugins.
• Hands-on experience configuring policy-driven security platforms such as SIEM, EDR, SOAR, DLP, CASB/SSE, or browser and endpoint agent security tools. 
• Proficiency with scripting or programming languages such as Python or PowerShell and experience working with REST APIs and webhook-based integrations.
• Strong written and verbal communication skills, with the ability to explain technical risk clearly to analysts and clients.
• Experience mentoring, training, or reviewing the work of analysts or junior engineers.

Preferred Qualifications

• Experience with AI security, agent runtime control 
• Familiarity with the OWASP Top 10 for LLM Applications, the OWASP Agentic Top 10, MITRE ATT&CK, MITRE ATLAS, NIST AI RMF, or ISO/IEC 42001.
• Experience administering or securing enterprise AI assistants, LLM platforms, or agent frameworks.
• Experience with enterprise EDR, SIEM, and cloud productivity security platforms.
• Relevant security, cloud security, or AI governance certifications such as Security+, CCSP, or IAPP AIGP.

Beazley Security Offers

• Competitive salary and bonus. 
• Flexible working arrangements. 
• Generous leave policies including 3 months paid parental.
• 100% of employee-only insurance premiums covered (healthcare, dental and vision).
• Up to 5% matched 401k contribution.
• Opportunities for career advancement and ongoing training.
• Participation in industry conferences and events.