LLM Security & Red Teaming
Break your AI application before attackers do.
Most engineers understand prompt injection in theory. Few have built the defenses. This course covers the OWASP LLM Top 10 with working exploit and mitigation code, indirect injection via RAG, supply chain attacks, excessive agency, insecure output handling, system prompt extraction, automated red teaming methodology, and a full defense-in-depth architecture. Capstone: security audit of a production RAG system with five planted vulnerabilities.
7h
Duration
8
Lessons
0
Learners
Course map
Lessons unlock as you complete the previous one. Your progress is saved on this device.
Lesson 1
The OWASP LLM Top 10: an engineer's tour
10m35 XPLesson 2
Prompt injection: direct, indirect, and multi-turn attacks
12m40 XPLesson 3
Supply chain attacks: poisoned RAG, tools, and models
10m35 XPLesson 4
Excessive agency and insecure output handling
11m38 XPLesson 5
System prompt extraction and information disclosure
9m33 XPLesson 6
Automated red teaming and building a security eval set
12m42 XPLesson 7
Defense in depth: a layered security architecture
10m35 XPLesson 8
Capstone: security audit of a production RAG system
18m60 XP
Take next
Courses that pair well after — or alongside — LLM Security & Red Teaming.