Skip to content

🗺️ The Big Map of AI (One-Page Overview)

⏱️ 8 min read🎯 Everyone🧰 Needs: nothing

Before diving into chapters, here's the whole territory on one page. Once you can see how models, apps, connectors, automations and your own data fit together, every chapter in this manual slots neatly into place.

✅ Key Points & Steps

The AI world can look chaotic, but it fits into five layers that work together. Knowing the layers helps you understand any new tool you come across.

  • Models are the AI systems that read and generate text, images and more.
  • Apps like ChatGPT, Gemini and Claude are how you talk to models.
  • Connectors, MCP and automations link those apps to your data and to other tools, so AI can act, not just answer.

🏙️ The whole landscape in one picture

flowchart TB
    YOU((🧑 You))
    subgraph APPS["🏢 Apps you talk to (hosts)"]
        A1["💬 Chat apps<br/>ChatGPT · Gemini · Claude · Copilot<br/>Grok · Perplexity · Meta AI"]
        A2[🛠️ Builder tools<br/>Claude Code · Codex · Cursor · VS Code]
        A3[🏡 Apps with AI inside<br/>Notion · Google · Microsoft · Obsidian]
    end
    subgraph BRAINS["🧠 Models (the brains)"]
        M1[☁️ Frontier models<br/>GPT · Gemini · Claude · Grok]
        M2[🏠 Open & local models<br/>Llama · Qwen · DeepSeek · Gemma · Mistral]
    end
    subgraph PIPES["🔌 Connections"]
        C1[MCP servers]
        C2[Built-in connectors & plugins]
        C3[APIs & webhooks]
    end
    subgraph ROBOTS["⚙️ Automation (runs without you)"]
        R1[n8n · Zapier · Make · Shortcuts]
    end
    subgraph STUFF["📦 Your stuff"]
        D1[Email · Calendar · Files]
        D2[Notes · Tasks · Code]
        D3[Smart home · Money · Media]
    end
    YOU --> APPS
    APPS --> BRAINS
    APPS --> PIPES
    ROBOTS --> BRAINS
    ROBOTS --> PIPES
    PIPES --> STUFF

🧱 The five layers

Layer What it is Examples Learn it in
🧠 Models The "brains" that read and write GPT, Gemini, Claude, Grok, Llama, DeepSeek, Mistral, Qwen Part III, Part IX
🏢 Apps (hosts) Where you talk to the brains ChatGPT, Gemini, Claude, Copilot, Grok, Perplexity, Meta AI, Cursor, Notion AI Part II, Part VI, Part VII
🔌 Connections How apps reach tools and data MCP servers, connectors, APIs Part IV
⚙️ Automation Workflows that run on triggers n8n, Zapier, Make, Shortcuts Part V
📦 Your data & knowledge What makes AI useful to you Files, notes, RAG, memory Part VIII

Everything else in the manual is about combining these layers: creative projects (Part X), real-life uses (Part XI), doing it well (Part XII), and full projects (Part XIII).

🪜 The skills ladder

flowchart LR
    L1[💬 1 · Chatter] --> L2[🔌 2 · Connector] --> L3[⚙️ 3 · Automator] --> L4[🛠️ 4 · Builder] --> L5[🏠 5 · Owner] --> L6[🏆 6 · Mentor]
Level You can… Next step
💬 1 · Chatter Ask questions, draft text Your First Hour
🔌 2 · Connector Let AI read and act on your apps MCP Explained
⚙️ 3 · Automator Make workflows that run without you Automation Platforms
🛠️ 4 · Builder Create apps, MCP servers and agents Agents & Coding Tools
🏠 5 · Owner Run private AI on your own hardware Local & Open Models
🏆 6 · Mentor Evaluate, secure, and teach others Teaching Others

❓ Which chapter answers my question?

Your question Go to
"I've never used AI. Where do I start?" Part I · AI from Zero
"Which assistant should I use: ChatGPT, Gemini, Claude…?" Meet the Assistants
"Why does the AI make things up?" How Models Really Work
"What's this MCP thing everyone talks about?" MCP Explained
"Which AI subscription should I pay for?" Choosing Your AI Stack
"Can AI handle my inbox?" Email & Calendar Superpowers
"How do I make AI do stuff on a schedule?" Automation Platforms
"Can I build an app with no coding experience?" Vibe Coding Your First Real App
"How do I make AI know my documents?" RAG, Memory & Knowledge
"Can I run AI without the cloud?" Local & Open Models
"How do I make a song or a video?" Music, Video & Audio
"Is my data safe?" Privacy & Your Data
"How do I avoid a surprise bill?" Cost Optimization
"Just give me a project to build!" The Build-Alongs

🔤 Your vocabulary starter pack

Word What it means
Model The AI system itself: a large pattern-learning program that predicts what text should come next
Token A chunk of text, roughly ¾ of a word. Models read, write and bill in tokens
Context window How much text the model can consider at once: your messages, files and its replies
Tool / function calling The model asking your app to run an action for it, like searching or creating an event
Agent An AI that works in a loop: decide on a step → use a tool → check the result → repeat until done
MCP An open standard that lets any AI app connect to any tool or data source
Connector A ready-made MCP connection inside an app, installed with a few clicks
RAG Retrieval-augmented generation: the AI searches your documents and answers from what it finds
Automation / workflow A rule that runs on its own: "when X happens, do Y"
Local model A model that runs on your own computer, privately and without an internet connection

🎯 Key takeaways

  • Five layers: models, apps, connections, automation, your data. Every AI setup is a combination of them.
  • The skills ladder runs from chatting to mentoring, and you can climb in any order.
  • Use the question table above as a shortcut to the right chapter.

🧠 Check yourself

❓ Is Cursor a model or a host?

A host (an app you use). It uses models like Claude or GPT under the hood.

❓ What's the difference between a connector and an automation?

A connector lets an AI app reach a tool when you ask. An automation runs on its own when a trigger happens (like a schedule or a new email).

🎮 Try this

Look at the landscape diagram and circle (mentally!) the pieces you already use. The empty spots are your adventure map. Pick one and open its part landing page from the sidebar.

Next up 01 · What Is AI, Really? →