Skip to content
anythingai.space
Conversational AIintermediate13 min · 9 steps

Conversational Design Patterns

Reusable shapes for chat systems: intent routing, generative fallback, hybrids, retrieval, tools, copilots and multi-agent teams, and when to pick each.

Intent-drivenroute by intentGenerative fallbackLLM when unsureHybridclassical + LLMCopilotAI inside the appMulti-agentspecialist agentsRetrieval-augmentedground in a KBTool-using agentsact via APIs
requestdataresponsetool call
1 / 9
Key ideaStep 1 of 9

Pick from a pattern library, not ad hoc

Design patterns are reusable answers to recurring problems in conversational systems. Building without one is risky. Forcing one pattern onto every use case rarely fits.

The recommended approach is a pattern library: know the main shapes and pick (or combine) the one that fits each job. Most production assistants mix several.

InsightThe first three patterns are about control: who decides the reply. The rest add capability (knowledge, actions) or change where the assistant lives.
What changed
  • +Seven patterns on the map

Lesson 7 of 8 in Conversational AI