Artificial Intelligence
Planning, tool use, memory and multi-agent systems.
Learn how to design tools, agents, state, routing, review, and human escalation with LangGraph—without requiring the article itself to execute.
The same Northstar Cloud incident-control problem, now combining LangGraph workflow control with MCP capability boundaries.
Build the Northstar Cloud multi-agent incident-control system with CrewAI agents and MCP servers, and compare it with the LangGraph/local-tool architecture from Part I.
Contrast fixed AI pipelines with systems that can choose actions, use tools, observe results and decide what to do next.
Turn the decision loop into a working agent that can choose a tool, observe its result, and decide when the task is complete.
How tool schemas connect model decisions to safe application functions, APIs, databases, and other capabilities.
How specialized agents can divide work, communicate through explicit artifacts, and collaborate without turning the system into unnecessary complexity.
Understand the control layer that coordinates models, tools, agents, state, retries, branching, and human checkpoints.
A clear mental model of agentic AI: the observe, reason, act loop, tool calling, and a minimal Python implementation.