Deep dive into red teaming LLMs, prompt injection defense, and enterprise AI safety frameworks.
Schedule
August 15 - August 22, 2026
Duration
2 weeks • 2hrs/wk
Project
Hands-on capstone
4 practical sections built around live exercises.
Identify the risks that make LLM and agent systems different from traditional apps.
Topics covered
Hands-on lab
Threat model a model-powered app and identify the highest-risk trust boundaries.
Test model systems and add layers that reduce real-world risk.
Topics covered
Hands-on lab
Run a small prompt-injection exercise and design mitigations.
Turn safety expectations into operational controls.
Topics covered
Hands-on lab
Draft a governance control map for one AI workflow.
Design a central layer for routing, monitoring, and policy enforcement.
Topics covered
Hands-on lab
Design a private LLM gateway architecture for an internal team.
Concrete capabilities you should leave with.
Threat model AI and agent systems
Understand prompt injection and RAG attack paths
Design layered safety controls
Create governance artifacts teams can actually use