AI Security ArchitectOversee AI architectural activities for a specific business or technology domain, and manage the development of solution architectures for projects or programs within a business area.Define AI security standards and direction of architecture in the specific business or technical domain, and establish best practices for protecting AI pipelines, datasets, and models.Define and develop the logical architectural design and strategies necessary to secure the Organizations AI domain/infrastructure.Utilize architecture patterns to suggest the most adequate utilization of technical platforms in support of the holistic AI solution security architecture design.Define, create and evolve the Architecture Governance Framework (e.g. architecture methods, practices and standards) for AI.Understand and advocate the principles of business and IT strategies, be prepared to sell the Architecture process, its outcome and ongoing results, and to lead the communication, marketing or educational activities needed to ensure Enterprise Architecture success and use.Assess the organization's AI landscape and identifying potential vulnerabilities or weaknesses including identification and evaluation of risks associated with training, deployment, and operation of AI models; keep up to date with the latest security threats, trends, and best practices to ensure the AI security infrastructure remains effective, and evaluate and select security tools, technologies, and products to enhance AI security.Collaborate with IT teams to integrate security measures into all aspects of the AI platforms and LLMs related processes, working with data scientists, engineers, and DevOps teams to embed security into the AI development lifecycle, and provide guidance and support to other Engineering teams in implementing security measures and resolving security related issues.Regularly reporting on the status of AI security measures to senior management and stakeholders.Securing AI systems from development through deployment, including securing training data and monitoring deployed models for threats. Knowledge of AI solutions development lifecycle and environments including MLOps and related tooling (e.g.
model repositories, data pipelines, deployment architectures).