Skip to main content
If you’ve been building AI agents with LangGraph or CrewAI — defining state graphs, tool nodes, conditional edges, and bolting on separate memory stores — this guide shows how Pixeltable replaces the graph DSL with declarative tables.
Related use case: Agents & MCP

Concept Mapping


Side by Side: Tool-Calling Agent

An agent that picks tools, calls them, and answers based on the results.
Packages: langgraph, langchain-openai, langchain-core, plus a vector DB client for RAG

What Changes


Common Patterns

Adding persistent memory

Adding RAG to an agent

Inspecting agent behavior


Next Steps

Agents & MCP

Full use case walkthrough

Agentic Patterns

All 8 agentic patterns as Pixeltable tables

Tool Calling

Register UDFs and queries as LLM tools

Pixelagent

Lightweight agent framework built on Pixeltable
Last modified on June 24, 2026