Advanced guides assume you already build with LLMs and want techniques beyond the basics. On the prompting side they cover methods from research papers, with working code and their measured limits: self-consistency and least-to-most, Self-Ask, Tree of Thoughts, Chain-of-Verification, semantic entropy, self-correction, active prompting, automatic prompt optimization, prompt ensembles, prompt compression and Skeleton-of-Thought.
On the engineering side they cover agent architecture: plan-and-execute and code-action agents, autonomous runtimes, self-modifying agents, multi-tenancy, agent identity and MCP authentication with OAuth. The security guides at this level cover red teaming, RAG security, supply chain trust and hardening model output.
The library lists all guides with filters for topic, level, category and reading time.