Modern websites, ranked in AI searchCited by ChatGPT, Perplexity & Google AI Overviews32% of our own traffic comes from AI searchModern websites, ranked in AI searchCited by ChatGPT, Perplexity & Google AI Overviews32% of our own traffic comes from AI searchModern websites, ranked in AI searchCited by ChatGPT, Perplexity & Google AI Overviews32% of our own traffic comes from AI searchModern websites, ranked in AI searchCited by ChatGPT, Perplexity & Google AI Overviews32% of our own traffic comes from AI search
Agentic AI

Agentic AI Design Patterns

Enterprise-focused guides to agentic AI — from prompt chaining and RAG to multi-agent systems and MCP. Based on Antonio Gulli's definitive guide.

All articles · 20

Agentic AI·March 23, 2026·11m

What Makes an AI System an Agent? The Foundation of Agentic Design

What makes an AI system a true agent, from the 5-step agentic loop to the 4 levels of agent complexity. Based on Antonio Gulli's Agentic Design Patterns.

Read more
Agentic AI·March 30, 2026·10m

Tool Use in AI Agents: Function Calling and Beyond

How AI agents use function calling to work with APIs and databases and other external services. The tool use pattern explained with code examples from Antonio Gulli.

Read more
Agentic AI·April 1, 2026·10m

Model Context Protocol (MCP): The New Standard for Agent-Tool Integration

The Model Context Protocol (MCP), the emerging standard for connecting AI agents to tools and data sources. From Antonio Gulli's Agentic Design Patterns.

Read more
Agentic AI·April 15, 2026·10m

RAG for AI Agents: Retrieval-Augmented Generation Done Right

Retrieval-Augmented Generation (RAG) for AI agents: how to ground agents in factual, up-to-date knowledge. Patterns and code from Antonio Gulli's Agentic Design Patterns.

Read more
Agentic AI·April 8, 2026·10m

Multi-Agent Systems: Orchestrating Teams of AI Agents

How to build multi-agent AI systems where specialized agents collaborate to solve complex problems. Orchestration patterns from Antonio Gulli's Agentic Design Patterns.

Read more
Agentic AI·June 15, 2026·9m

Prompt Chaining: Building Reliable AI Agent Workflows

Prompt chaining breaks a complex task into a sequence of LLM calls, where each step's output feeds the next. It is the foundational pattern for reliable AI agents.

Read more
Agentic AI·June 15, 2026·10m

Routing and Parallelization: Scaling AI Agent Orchestration

Routing dispatches each input to the right specialized path; parallelization runs independent sub-tasks at once. Together they scale agent accuracy and latency.

Read more
Agentic AI·June 15, 2026·10m

Reflection and Adaptation: How AI Agents Learn From Their Own Output

Reflection is the pattern where an AI agent critiques its own output and revises it, looping until the work clears a quality bar. It is the self-correction loop behind reliable agents.

Read more
Agentic AI·June 15, 2026·10m

AI Agent Planning: How Intelligent Systems Decide What to Do Next

AI agent planning turns a high-level goal into an ordered sequence of executable steps. Learn task decomposition, plan-and-execute vs ReAct, and when to replan.

Read more
Agentic AI·June 15, 2026·11m

Reasoning Techniques for AI Agents: Chain-of-Thought to Tree-of-Thought

Chain-of-Thought, self-consistency, ReAct, and Tree-of-Thought are the reasoning architectures that separate toy AI demos from production agents.

Read more
Agentic AI·June 15, 2026·10m

Agent-to-Agent Communication: How A2A Enables Agent Interoperability

A2A (Agent2Agent) is an open protocol that lets agents from different vendors discover each other, advertise capabilities, and delegate tasks across systems.

Read more
Agentic AI·June 15, 2026·11m

Memory Management for AI Agents: Short-Term, Long-Term, and Beyond

How AI agents remember across a turn and across sessions: short-term context, long-term episodic, semantic, and procedural memory, and how it differs from RAG.

Read more
Agentic AI·June 15, 2026·11m

AI Guardrails and Safety: Building Trustworthy Agentic Systems

AI guardrails are the input, output, and permission controls that keep an agent safe in production: what separates a demo from an enterprise deployment.

Read more
Agentic AI·June 15, 2026·10m

Exception Handling and Human-in-the-Loop: Making AI Agents Resilient

How AI agents fail gracefully: retries, fallbacks, and circuit breakers for tool failures, plus human-in-the-loop approval gates for high-risk actions.

Read more
Agentic AI·June 15, 2026·12m

Monitoring AI Agents: Goal Setting, Evaluation, and Prioritization

You can't improve an AI agent you can't measure. A practical guide to observability, offline and online evals, the metrics that matter, and what to fix first.

Read more
Agentic AI·June 15, 2026·10m

Resource-Aware AI Agents: Optimization and Exploration Strategies

Resource-aware AI agents budget tokens, route tasks by difficulty, cache results, and stop early. The patterns that separate a viable product from a money pit.

Read more
Agentic AI·June 15, 2026·10m

Industry Leaders on Agentic AI: Perspectives from Google and Goldman Sachs

What Google's Saurabh Tiwary and Goldman Sachs' Marco Argenti reveal about how enterprises build, govern, and get ROI from agentic AI.

Read more
Agentic AI·June 15, 2026·11m

The Agentic AI Toolkit: Frameworks, Environments, and CLI Agents

A working tour of the agent-building stack: orchestration frameworks (LangGraph, CrewAI, AutoGen, ADK), agent platforms, and CLI coding agents, and when to reach for each.

Read more
Agentic AI·June 15, 2026·16m

The Definitive Glossary of Agentic AI: 100+ Terms Explained

A 100+ term glossary of agentic AI, from agents and the agentic loop to MCP, RAG, reflection, guardrails, and evals. Clear definitions for teams building agents.

Read more