Agent Design Patterns (21)
A distilled take on the 21 core patterns from Google's Agentic Design Patterns. A "which pattern for which problem" cookbook that fills the blind spots of the first three modules.
The first three modules cut along "capability dimensions," "engineering foundations," and "harness methodology." This module picks up a different knife — cutting along 21 reusable implementation patterns, a cookbook of "which pattern for which problem."
The source is Google's Agentic Design Patterns (Antonio Gulli); this is a distilled Chinese-to-English rendering, referencing xindoo's translation project. The 21 patterns group into 6 clusters: core execution / quality & safety / memory & knowledge / collaboration & interop / control & governance / reasoning & exploration.
Lecture order: Lecture 0 is the panorama — laying out all 21 patterns at once, grouping them, and marking "covered in the first three modules or not"; the later lectures dive only into the blind-spot patterns (reflection, learning & adaptation, knowledge retrieval / RAG, agent-to-agent communication / A2A, reasoning techniques, prioritization & exploration), without repeating tools/planning/memory/exception/safety/evaluation already covered.
For: anyone who wants a "move cheat-sheet" — knowing, for a given class of problem, which established patterns are on the menu.
Posts
A Panoramic Map of the 21 Agent Design Patterns
Lay out all 21 patterns from Google's "Agentic Design Patterns" at once, group them into 6 clusters, and mark which ones this course has already covered.
D · 1 Patterns · Post 1"Reflection and Reasoning: Letting an Agent Go Back and Fix Its Work, and Think More Deeply"
A one-pass output can be wrong without the Agent knowing it; Reflection adds critique-driven, targeted revision; Reasoning makes the thinking explicit so the Agent thinks more deeply.
D · 2 Patterns · Post 2"Learning & Adaptation + RAG: So Your Agent Stops Starting From Scratch — and Stops Making Things Up"
Learning lets an Agent improve its strategy from feedback; RAG lets it plug into your private knowledge — both work at the context layer.
D · 3 Patterns · Post 3"Agent-to-Agent Communication (A2A): When Agents Start Talking, the Message Deforms"
The real hard part of multi-agent collaboration isn't the division of labor — it's how they pass messages without distortion.
D · 4 Patterns · Post 4"Prioritization + Exploration: Which to Do First, and How to Avoid Always Taking the Beaten Path"
Prioritization answers "which to do first"; exploration answers "do you dare walk an untrodden path"—the finale of Module D.