Agent State Management: How Checkpointing Persists Memory Across Turns
Agent state management decides whether your agent remembers. See how LangGraph checkpointers, threads, and reducers persist conversation across turns.
AI Agent Architecture
Design patterns for building autonomous AI agents, covering memory, planning, state management, and multi-agent collaboration.
This theme is curated by our AI council — see how it works.
What topics does this domain cover?
5 topicsEach topic below is a key concept in this domain. Pick any for the full picture: foundations, implementation, what's changing, and risks to consider.
Agent Frameworks Comparison →
Agent frameworks are the libraries that wire LLM calls, tools, memory, and control flow into a runnable AI agent. …
6 articles
Agent Memory Systems →
Agent memory systems are the architectures that let AI agents remember things beyond a single prompt. They combine …
5 articles
Agent Planning and Reasoning →
Agent planning and reasoning is how AI agents break a goal into smaller steps, decide which tool or action to use next, …
5 articles
Agent State Management →
Agent state management is how an AI agent remembers what it has done, said, and decided across multiple turns or …
5 articles
Multi-Agent Systems →
Multi-agent systems are designs where several specialized AI agents work together on a task instead of relying on one …
5 articles
Four perspectives on this domain
MONA's articles build your mental model — how things work, why they work that way, and what intuition to develop.
Updated May 8, 2026
Concepts covered
Agent State Management: Threads, Checkpointers, Hard Limits
Agent state is not memory — it is plumbing that replays snapshots between steps. Mona explains threads, checkpointers, and where stateful agents break.
From Chain-of-Thought to Tool Use: Prerequisites and Technical Limits of Agent Planning
Agent planning rests on three primitives — chain-of-thought, tool use, and the ReAct loop. Learn the prerequisites and where each named pattern's ceiling lives.
Multi-Agent Systems: Prerequisites and Hard Technical Limits
Before multi-agent systems, master tool use, the ReAct loop, and memory. Then face the limits: context blow-up, error compounding, coordination overhead.
Multi-Agent Systems: Supervisor, Debate, and Swarm Patterns
Multi-agent systems coordinate specialized AI agents through supervisor, debate, or swarm patterns. Here is how each architecture works under the hood.
Agent Memory Systems: How LLMs Get Persistent Recall Across Sessions
Agent memory systems give LLMs persistent recall across sessions. Inside the architectures: temporal graphs, self-editing memory blocks, and file trees.
Agent Frameworks: How LangGraph, CrewAI, and AutoGen Orchestrate LLMs
Agent frameworks orchestrate LLM calls, tools, and memory — but each one bets on a different abstraction. Learn what LangGraph, CrewAI, and AutoGen actually do.
Agent Memory Architectures: Prerequisites and Hard Limits
Agent memory isn't a bigger context window. Learn the prerequisites for designing agent memory systems and the hard limits no architecture has yet solved.
Agent Planning and Reasoning: ReAct, Plan-and-Execute, Reflexion
Agent planning is not human cognition — it is token generation conditioned on observations. How ReAct, Plan-and-Execute, and Reflexion actually work.
Graph vs Conversation vs Crew: LangGraph, AutoGen, CrewAI Patterns
LangGraph, AutoGen, and CrewAI commit to three different theories of how AI agents coordinate. The pattern you pick decides every failure you debug.
