Agentic AI Roadmap

Agentic AI Roadmap — a step-by-step learning path covering Foundations of Agentic AI, LLM & Prompting Fundamentals, Tool Use & Function Calling, Retrieval…

← All skill roadmaps · Create a free account to track your progress

Foundations of Agentic AI

What AI agents are, how they differ from chatbots and workflows, and where they shine.

• Anthropic – Building Effective Agents
• Hugging Face Agents Course
• Microsoft – AI Agents for Beginners Course
• freeCodeCamp – AI Agents for Beginners
• freeCodeCamp – How to Build Agentic AI Workflows
• Google – Agents Whitepaper

LLM & Prompting Fundamentals

The language-model core every agent is built on.

• The Illustrated Transformer
• Prompt Engineering Guide
• Andrej Karpathy – Intro to Large Language Models
• Claude Models Overview

Tool Use & Function Calling

Giving agents the ability to act on the world through tools and APIs.

• Claude Tool Use Documentation
• OpenAI Function Calling Guide
• DeepLearning.AI – Functions, Tools and Agents with LangChain
• LangChain Tools Concepts

Retrieval-Augmented Generation

Grounding agents in private or fresh data with retrieval.

• Anthropic – Introducing Contextual Retrieval
• LangChain RAG Tutorial
• DeepLearning.AI – Building Agentic RAG with LlamaIndex
• Pinecone – What is RAG?

Planning & Reasoning

How agents break down goals and decide what to do next.

• ReAct: Synergizing Reasoning and Acting in LLMs
• Reflexion: Language Agents with Verbal Reinforcement Learning
• DeepLearning.AI – AI Agentic Design Patterns with AutoGen
• LangGraph Agent Architectures

Memory & State Management

Giving agents continuity across steps, sessions, and users.

• LangGraph Memory Concepts
• MemGPT: Towards LLMs as Operating Systems
• DeepLearning.AI – LLMs as Operating Systems: Agent Memory
• Mem0 – Memory Layer for AI Agents

Model Context Protocol (MCP)

The open standard for connecting agents to tools, data, and services.

• Model Context Protocol – Introduction
• DeepLearning.AI – MCP: Build Rich-Context AI Apps with Anthropic
• Microsoft – MCP for Beginners
• MCP Servers Repository
• Anthropic – Introducing the Model Context Protocol

Multi-Agent Systems

Orchestrating teams of specialized agents to solve bigger problems.

• LangGraph Multi-Agent Systems
• DeepLearning.AI – Multi AI Agent Systems with crewAI
• freeCodeCamp – Agentic AI with LangGraph (24h course)
• CrewAI Documentation
• OpenAI Agents SDK
• AutoGen (Microsoft)

Evaluation & Observability

Measuring, tracing, and debugging agent behavior in the real world.

• Anthropic – Demystifying Evals for AI Agents
• DeepLearning.AI – Evaluating AI Agents
• LangSmith Observability
• Langfuse – Open-Source LLM Engineering

Safety, Guardrails & Production

Shipping agents that are safe, reliable, and cost-effective.

• OWASP Top 10 for LLM Applications
• Gandalf – Prompt Injection Game (Lakera)
• OpenAI – A Practical Guide to Building Agents
• Simon Willison – Prompt Injection Series