This course is designed for IT professionals, security practitioners, SOC analysts, and developers who want to build a practical understanding of how modern AI and LLM-based systems work from a security perspective. It is ideal for those with basic knowledge of applications and security who need to analyze AI-enabled systems with confidence, without a background in machine learning or data science. It starts with the fundamentals of AI and LLMs, including prompts, tokens, context windows, and inference, before progressing into real-world system architectures such as model endpoints, orchestrators, agents, and RAG pipelines. Learners will analyze how data flows through these systems, identify trust boundaries, and examine where sensitive data can be exposed through prompts, retrieval systems, logs, and vector databases. The course will then conclude with practical analysis techniques that enable learners to inspect architectures, trace data flows, and identify potential security risks in deployed AI applications.
