Agents
Models that reason, call tools, and act across multiple steps.
8 Agents
- What Is an AI Agent? Strip the hype and an agent is three things: a model that can call tools, a loop, and the authority to decide what happens next.
- Agent Architectures: From Rules to Autonomy A spectrum runs from fixed pipelines to fully autonomous loops. Knowing where a task belongs on it is the single most important agent design decision.
- How Does Tool Calling Work? The model never runs anything. It emits a structured request, your code executes it, and the result goes back as text. Everything follows from that.
- The Agent Loop Think, act, observe, repeat. The cycle is trivial to implement and the hard parts are all about when to stop.
- How Agents Fail Seven recurring failure modes, what causes each, and the specific guardrail that prevents it. Read this before shipping anything autonomous.
- Writing Good Tool Definitions The tool description is a prompt, and it is the one the model relies on most. Vague descriptions cause more agent failures than weak reasoning.
- What Is MCP? A standard protocol for exposing tools to models, so integrations stop being written once per application.
- Multi-Step Planning Plan upfront or decide as you go? The tradeoff is coherence against adaptability, and the practical answer is usually a shallow plan you revise.
- The ReAct Pattern Interleave reasoning and action instead of separating them. The idea behind the modern agent loop, why it beats reasoning alone, and where it strains.
14 Advanced RAG
15 Advanced Agents
- Evaluating Agents Score the outcome, not the path. Agents are nondeterministic and multi-step, which breaks the evaluation habits that work for single calls.
- Agent Memory Models have no memory between calls. Everything that feels like memory is retrieval, and designing what to store is the actual problem.
- Coding Agents The agent application that works best in practice, and the specific properties of software work that make it work.
- Computer-Use Agents Agents that see a screen and click. Universal compatibility, unreliable execution, and a security model that demands isolation.
- Context Engineering Deciding what goes into the context window, in what order. The discipline that replaced prompt engineering once systems got complicated.
- Human-in-the-Loop Design Where to put a person in an automated system. Gate by reversibility and confidence, not by task importance.
- Multi-Agent Systems Several agents with different roles, coordinating. Sometimes better than one, often just a more expensive way to fail.