AI Agents
From "what is an agent?" to multi-agent orchestration and MCP.
Agents are LLMs that do things: they perceive, plan, call tools, and act with some autonomy. This guide orders every agent article on this site into one path — starting with what actually makes something an agent (and what's just marketing), through the reasoning and prompting techniques agents run on, to memory, the Model Context Protocol, and coordinating whole teams of agents.
Start here
What an agent is — and isn't.
01What is an AI Agent?
An LLM becomes an agent when it can reason about which tool to call, execute that call, and update its plan based on the result. The actual loop running underneath every agent f...
02Agentic AI: From Passive Models to Autonomous Systems
Agentic AI isn't just a smarter chatbot, it's an LLM wired to tools, memory, and a feedback loop. Breaking down the core components before the framework abstractions take over.
How agents think and act
The loop inside every agent: reason, decide, act.
03Agentic AI: How Autonomous Systems Perceive, Plan, and Act
Most AI systems react to input. Agentic systems plan, take actions, and recover from errors autonomously. The architecture, perception, memory, reasoning, action, that makes the...
04Advanced Prompt Engineering: Chain-of-Thought, ReAct, and Tree-of-Thoughts Explained
Chain-of-thought improves multi-step reasoning. ReAct adds tool use. Tree-of-thoughts explores multiple solution paths. When each technique earns its token cost, and when simple...
05Context Engineering: The New Skill That Is Replacing Prompt Engineering
Prompt engineering is giving way to something deeper: context engineering. How you structure what goes into the context window, what you include, what you leave out, and in what...
06Structured Outputs in LLMs: JSON Mode, Function Calling, and Schema Validation
Free-form LLM output breaks parsing pipelines. JSON mode, function calling, grammar-constrained decoding, and Pydantic validation are the layers that make structured output reli...
Memory & tools
What an agent remembers, and how it reaches the outside world.
07AI Agent Memory: Short-Term Context, Long-Term Storage, and Episodic Recall
Stateless LLMs forget everything when the context window closes. Building agents that actually remember requires understanding four distinct memory types and when to use each one.
08Model Context Protocol (MCP): A Complete Beginner's Guide
MCP is the USB-C port for AI applications, one protocol that connects any LLM host to any external tool or data source. This guide covers the architecture, three core primitives...
Scaling up
Many agents, real SDKs, and multimodal deployment hurdles.
09Multi-Agent Systems: Orchestration, Communication, and Collaborative AI
A single agent hits context and capability limits fast. Multi-agent systems distribute work across specialized roles with structured communication protocols. Orchestration patte...
10OpenAI Agents SDK vs Anthropic SDK: A Technical Comparison
OpenAI and Anthropic both now ship production-ready agent frameworks. This post compares them side by side: how each models an agent, handles tool calls, orchestrates multi-agen...
11Navigating the 3 Critical Hurdles of Multimodal AI Agent Deployment
Multimodal agents hit three hard walls in production: image token cost, latency from vision encoding, and grounding errors that compound across reasoning steps. How to engineer ...
Tools & resources
Free utilities that pair with this guide.
Keep going
Guides that pick up where this one ends.