Guardrails and AI: Building Safe + Controllable Apps

In Chapter 3, you combined all your learnings on RAG, memory, and embeddings with tools and chains to create an end-to-end agent – that could plan out your day for you. This agent was able to reason and have access to “the world” via API integrations (the so-called tools). This was a fairly simple application, but it was still autonomous – and when AI is autonomous, there’s always space for things to go wrong if proper safeguards are not in place.