AI Security Architect

Synopsys•Morrisville, NC

3d•$201,000 - $301,000•Hybrid

About The Position

Synopsys is seeking a highly motivated Cybersecurity Architect - AI Security to help define, architect, and secure the next generation of AI-enabled systems, platforms, and services across Synopsys. The Cybersecurity Architect - AI serves as a senior technical leader responsible for developing and implementing security architectures, standards, and governance frameworks for Artificial Intelligence (AI), Generative AI, Agentic AI, Machine Learning (ML), and Large Language Model (LLM) ecosystems. This role involves shaping how a global technology leader secures AI at scale, across products, platforms, and operations that power the semiconductor industry.

Requirements

Bachelor's degree in Computer Science, Cybersecurity, Engineering, Information Systems, or related field, Master's degree preferred.
10+ years of experience in Cybersecurity, Security Architecture, Cloud Security, or related disciplines with a track record of building and securing complex systems.
3+ years of hands-on experience securing AI, Generative AI, LLMs, or ML systems in production environments, not just research or proof-of-concept work.
Deep technical knowledge of AI and ML security, LLM architectures, Agentic AI systems, RAG pipelines, AI model lifecycle security, AI governance frameworks, cloud security architectures, IAM, API security, data security and privacy, secure software development lifecycle, and security monitoring and incident response.
Experience conducting threat modeling, architectural risk assessments, and security design reviews for AI systems, with the ability to identify and mitigate risks like prompt injection, data poisoning, model theft, and agent abuse.
Demonstrated experience working with executive stakeholders and cross-functional teams to drive security strategy, present risk and mitigation recommendations, and influence decision-making at the highest levels.
Strong familiarity with Kubernetes and container security, MLOps and ModelOps platforms, AI orchestration frameworks, vector databases, and the Model Context Protocol (MCP), as well as leading AI ecosystems including Anthropic, OpenAI, NVIDIA AI, Azure OpenAI, AWS Bedrock, and Google Vertex AI.
CISSP, CCSP, SABSA, TOGAF, or equivalent certifications, GenAI security-related certifications are a strong plus.

Nice To Haves

You can explain a complex AI security tradeoff to a VP in two sentences without losing the nuance, and then turn around and whiteboard the technical architecture with an AI engineer.
You do not wait for perfect information or fully mature frameworks, you work with what you have, ask the right questions, build what is needed, and iterate as the landscape evolves.
You push back when a design is too open or a risk is being underestimated, and you do it in a way that moves the conversation forward rather than shutting it down.
You are organized enough to manage multiple AI security initiatives at once, from architecture reviews to policy development to vendor evaluations, without losing track of what matters most.
You stay current on emerging AI threats, attack techniques, and defensive technologies, not because it is required but because you are genuinely curious about how these systems break and how to make them more resilient.
You have worked in semiconductor, software, SaaS, or technology industries where intellectual property protection and customer trust are non-negotiable, and you understand what is at stake when AI security fails.

Responsibilities

Design enterprise AI security architecture patterns and reference architectures for GenAI, Agentic AI, ML, and LLM-based systems across cloud, SaaS, on-premises, and hybrid environments.
Conduct AI threat modeling, architecture reviews, and security assessments for AI use cases, evaluating risks related to prompt injection, data poisoning, model theft, model inversion, membership inference, and agent abuse.
Define security standards and implementation patterns for agentic AI platforms, orchestration frameworks, MCP ecosystems, LLM gateways, and autonomous workflows, including policy enforcement, monitoring, logging, and action controls.
Establish controls for secure model training, fine-tuning, deployment, and monitoring, and drive AI red-team and adversarial testing strategies to validate defenses.
Design and implement Agentic Identity and Access Management (IAM) strategy for AI agents, autonomous systems, MCP servers, and AI-enabled applications, including controls for service access, secrets protection, model registries, vector databases, RAG architectures, API security, and data classification.
Lead secure AI and model lifecycle initiatives by defining security controls, architectures, and guardrails for AI platforms, agents, and LLM integrations; implementing protections against AI-specific threats; and driving red-team testing and governance to ensure secure, resilient, and trustworthy AI systems.
Partner with Product Security, Engineering, Cloud, IT, Legal, Privacy, Data Governance, and Enterprise Architecture teams to establish AI security policies, standards, and governance frameworks aligned with NIST AI RMF, ISO 42001, OWASP LLM Top 10, and emerging regulations.
Research emerging AI threats and defensive technologies, evaluate and implement AI Security Posture Management (AI-SPM) tools, and design secure sandbox environments for rapid security evaluation of new AI models, solutions, and agents.