01
intermediate45 min
What Is an AI Agent ; The Loop, Memory, and Tools
The architectural reality of AI agents: the ReAct loop, tool use, memory layers, and why 'agent' means something specific and important.
agentreacttoolsmemoryloop
Build real agents: the ReAct loop, tool use, memory systems, and the architectural primitives that turn an LLM into an autonomous actor.
The architectural reality of AI agents: the ReAct loop, tool use, memory layers, and why 'agent' means something specific and important.
Hands-on: wire up a ReAct agent from scratch using the Anthropic SDK. Tool definitions, the message loop, state handling, error recovery.
How agents remember things: context window tricks, external key-value stores, vector databases, and the architectural decisions that determine what your agent knows.
How tools extend agent capabilities: schema design, parameter validation, dangerous tools, and building tool suites for offensive workflows.