Generative AI Security Engineer – Agentic AI & Secure SDLC Automation

Iconma Portal•Phoenix, AZ

2d•Hybrid

About The Position

Our client, an IT Services and Consulting company, is seeking a Generative AI Security Engineer specializing in Agentic AI and Secure SDLC Automation for a hybrid role in Phoenix, Arizona. The position serves as a subject matter expert in generative AI and agentic engineering, building and operationalizing production-grade autonomous and semi-autonomous AI agents that support secure, scalable enterprise automation. The engineer will develop security automation tooling, integrate security controls into pull request and CI/CD workflows, and build agents that embed security throughout the software lifecycle. The role includes designing secure integrations with enterprise systems and developer tools, applying controls such as least privilege and secrets protection, and establishing practices to evaluate agent quality, security, reliability, latency, and cost. It also provides technical leadership and mentorship in agent architecture, AI-assisted engineering, secure coding, and production readiness, while supporting the security champion practice.

Requirements

Senior-level technical competence and leadership capability, serving as a subject matter expert in generative AI and agentic engineering.
10+ years of software engineering experience.
10+ years of experience with automation and development in CI/CD.
At least 5 years of cybersecurity experience.
At least 2 years of generative AI experience.
Hands-on experience delivering agentic AI or AI coding-agent solutions into production, preferably in security, developer productivity, or Secure SDLC workflows.
Demonstrated experience designing, building, and productionizing autonomous or semi-autonomous AI agents, including planning, tool use, memory and context management, multi-agent orchestration, human-in-the-loop controls, and failure recovery.
Hands-on experience with AI coding and agent platforms such as Anthropic Claude/Claude Code and OpenAI Codex, including prompt and instruction design, tool integration, API/SDK usage, evaluation, guardrails, and enterprise adoption patterns.
Experience creating reusable agent tools, skills, workflows, and orchestration patterns using frameworks such as LangGraph, AutoGen, CrewAI, custom orchestration, and tool-protocol patterns such as Model Context Protocol (MCP).
Experience building automation tooling for security or developer efficiency.
Proficiency in at least one scripting language: Python, JavaScript, or Bash.
Familiarity with at least one compiled language: Go, Rust, or Java.
Experience with GitHub automation and API integrations.
Strong understanding of OWASP Top 10 and industry best practices in generative AI security.
Knowledge of agent frameworks and AI lifecycle management, including model integration, evaluation, and observability.
Deep expertise in security automation, CI/CD and pull request integrations, secure-by-design AI architectures, and embedding security across the software lifecycle, including threat modeling, secure code reviews, and vulnerability analysis.
Familiarity with Oracle, SQL, and NoSQL database architectures.
Experience mentoring teams in application security and secure development practices.
Strong knowledge of adversarial robustness techniques and tools for machine learning.
Strong knowledge of AI risk management frameworks and trustworthy AI practices.
Bachelor’s degree in Cybersecurity, Data Science, Computer Science, or Software Engineering.

Nice To Haves

Master’s degree in Cybersecurity, Data Science, Computer Science, or Software Engineering.
Information security professional certifications, such as SANS GIAC, CISSP, or generative AI certifications.
Machine Learning Operations professional certifications.

Responsibilities

Develop and maintain security automation tooling to improve visibility, enforce security best practices, and streamline processes.
Automate secure design reviews and integrate security checks into pull request workflows and CI/CD pipelines.
Design, develop, and operationalize agentic AI systems that can plan, execute, and adapt tasks across complex workflows.
Integrate agents with enterprise systems, APIs, data sources, and security platforms.
Build Secure SDLC agents that support requirements and secure design review, threat modeling, code review, dependency and vulnerability analysis, remediation guidance, pull-request checks, and release/security gates.
Design agents to interact securely with source control, CI/CD, issue tracking, security scanners, cloud services, internal APIs, and developer tooling.
Enforce least privilege, secrets protection, authorization boundaries, auditability, and approval workflows in agent interactions.
Establish agent evaluation practices covering task success, groundedness, code quality, secure coding, prompt-injection resistance, data leakage, unsafe tool use, permissions, reliability, latency, and cost.
Apply secure-by-design engineering to AI-assisted development, including sandboxing, policy enforcement, telemetry and observability, traceability, red-team testing, and controls for agent-generated code before merge or deployment.
Provide technical leadership and mentorship in agent architecture, AI-assisted engineering, secure coding, and production readiness.
Translate emerging agent capabilities into scalable enterprise engineering patterns.
Automate integrations and contribute to managing, monitoring, and reporting on generative AI and agentic AI security risk reduction.
Conduct research on emerging threats, vulnerabilities, and technologies to stay current with the security landscape.
Perform proofs of concept involving security tooling, zero-day vulnerabilities, and emerging risks.
Support the security champion practice by promoting security principles and controls.