Role Summary

We are seeking a Senior AI Security Engineer to own and advance the security posture of our AI-powered products, platforms, and infrastructure. As organizations rapidly adopt LLMs, agentic AI systems, and AI-augmented workflows, the attack surface has fundamentally shifted — and so must our security approach. You will operate at the intersection of cybersecurity and artificial intelligence, defending against novel AI-specific threats while enabling engineering teams to ship AI features quickly and safely.

This role spans the full AI security lifecycle: from threat modeling LLM integrations and designing guardrails against prompt injection, to securing model supply chains, hardening RAG pipelines, and building automated security tooling that scales with our AI platform. You will be the team's go-to authority on AI security risks, responsible AI safeguards, and emerging compliance requirements including the EU AI Act.

Key Responsibilities

AI/LLM Security Architecture & Engineering

• Design and implement security architectures for LLM-powered applications, AI agents, and copilot experiences;
• Build guardrails and defensive layers against AI-specific attack vectors: prompt injection (direct and indirect), jailbreaking, data poisoning, model inversion, membership inference, and training data extraction;
• Architect secure RAG (Retrieval-Augmented Generation) pipelines, ensuring data isolation, access control enforcement, and context boundary integrity;
• Implement output filtering, content safety classifiers, and toxicity detection for all AI-generated content;
• Design and enforce authentication, authorization, and rate-limiting for AI/LLM API endpoints and agentic tool-use interfaces;
• Secure model serving infrastructure including input/output logging, audit trails, and anomaly detection.

AI Threat Modeling & Red Teaming

• Conduct AI-specific threat modeling for all new AI features and integrations using frameworks such as STRIDE, MITRE ATLAS, and OWASP LLM Top 10;
• Lead and participate in AI red team exercises: adversarial prompt testing, model robustness evaluation, and data exfiltration simulations;
• Develop and maintain an AI threat intelligence capability — tracking emerging attack techniques, CVEs in AI/ML frameworks, and adversarial research publications;
• Build and maintain automated adversarial testing suites for continuous security validation of LLM integrations;
• Evaluate third-party AI models, APIs, and SaaS tools for security risks before organizational adoption.

AI Governance, Compliance & Responsible AI

• Drive compliance with AI-specific regulations: EU AI Act, NIST AI RMF, ISO 42001, and industry-specific guidance;
• Define and enforce AI security policies covering model access, data handling, prompt logging, and output monitoring;
• Collaborate with legal, compliance, and privacy teams on AI data governance, consent management, and bias auditing;
• Implement AI model cards, risk assessments, and documentation standards for all deployed models;
• Monitor and enforce responsible AI principles: fairness, transparency, explainability, and human oversight.

Security Platform & Tooling

• Build and maintain AI security tooling: automated prompt injection scanners, model vulnerability scanners, and AI-specific SAST/DAST rules;
• Implement AI-aware security monitoring and alerting in SIEM/SOAR platforms;
• Develop security guardrails as reusable libraries and middleware that engineering teams can adopt with minimal friction;
• Create security-as-code patterns for AI deployments: policy-as-code (Open Policy Agent, Cedar), infrastructure security scanning, and secrets management;
• Instrument AI systems with security telemetry for real-time threat detection and forensic analysis.

Cross-Functional Security Partnership

• Embed with AI engineering teams during design and development to ensure security is built-in, not bolted-on;
• Provide security architecture guidance to product managers, AI engineers, and data engineers on secure AI system design;
• Lead security training and awareness programs focused on AI-specific threats for engineering teams;
• Participate in incident response for AI/ML-related security events including model compromise, data leaks, and adversarial attacks;
• Serve as the internal subject matter expert on AI security, presenting at team tech talks and contributing to security documentation.