AI Security Engineer III

Overview This role reports into Director/Sr. Manager within Technology GRC and is the technical builder behind the function's agentic AI transformation. The AI Security Engineer III designs, builds, and operates the AI agents and automations that deliver GRC services - with primary focus on the GRC Tool platform, Technology Risks, Threats and Controls Library, the Risk Register, and Policy Governance - all operating under a human-in-the-loop model supervised by the responsible GRC leader.

The role bridges GRC domain requirements and engineering execution, prototyping control automations, evidence collectors, and governance tooling that reduce manual effort and enable the team to scale without proportional headcount growth.

• Bachelor's degree in Computer Science, Software Engineering, Data Science, or a related field; equivalent practical experience considered.
• Minimum 4+ years in software/AI engineering, with hands-on experience building LLM-powered applications and agentic workflows.
• Proficiency in Python (or comparable) and modern AI development tooling (e.g., Claude Code, Cursor, GitHub Copilot).
• Hands-on experience with LLM application patterns - prompt engineering, RAG, function/tool calling, agentic orchestration, and MCP.
• Familiarity with leading LLMs (Anthropic Claude, OpenAI GPT/o-series, Google Gemini, Meta Llama, Mistral) and model selection trade-offs (reasoning depth, context window, cost, latency, data residency).
• Working knowledge of the AI/LLM risk landscape: OWASP Top 10 for LLM Applications, NIST AI RMF, ISO/IEC 42001, MITRE ATLAS, and emerging regulation (EU AI Act, NYDFS AI guidance).
• Experience integrating with enterprise platforms and APIs; familiarity with GRC tooling (e.g., ServiceNow IRM, Archer, AuditBoard) a plus.
• Ability to translate GRC domain requirements into well-governed, production-grade automations with human-in-the-loop controls.
• Strong collaboration skills and the ability to partner with non-technical GRC stakeholders.
• Experience building agent evaluation frameworks, guardrails, prompt/version management, and observability/logging for production LLM systems.
• Familiarity with cloud ML platforms (AWS Bedrock/SageMaker, Azure AI, GCP Vertex) and CI/CD-integrated deployment gates.
• AI governance certification a plus (e.g., IAPP AIGP), including agentic architecture concepts.
• Preferred experience in the Property Management, Multifamily Housing, SaaS, FinTech, or PropTech industries.

• Build and integrate agentic AI solutions and automations within the enterprise GRC platform to standardize workflows, automate evidence collection, and improve reporting.
• Develop control automations, evidence collectors, and governance tooling, reducing dependence on the broader engineering backlog.

• Engineer AI-assisted workflows to populate, maintain, and reconcile the enterprise Risk Register across technology domains.
• Automate linkage between risk entries, controls, and remediation tracking.

• Build AI-assisted tooling for policy drafting, framework crosswalks, annual review cadence, and exception workflows.

• Design agentic workflows using RAG, function/tool calling, and Model Context Protocol (MCP), with appropriate guardrails for accuracy, confidentiality, and IP boundaries.
• Implement human-in-the-loop checkpoints and monitoring so GRC leaders can supervise and validate agent outputs.
• Apply AI risk controls aligned to OWASP Top 10 for LLM Applications, NIST AI RMF, ISO/IEC 42001, and MITRE ATLAS.

• Build evaluation harnesses and observability for deployed agents - measuring grounding, accuracy, and consistency while minimizing hallucinations across GRC use cases.
• Implement guardrails, deployment gates, and immutable audit trails/logging so non-compliant or low-confidence outputs are caught before use.
• Maintain model documentation (model cards, data provenance) to support AI governance and regulatory defensibility.