What Is a Context Window? Why Long Chats Start Forgetting
A long conversation with an AI starts strong and then quietly stops holding on to what you agreed. People read that as the model losing the thread. It is simpler than that.…
Visual explainers • nine panels each • free
Neural
Neural
Canvas
One idea, drawn nine ways. Built to make sense with no caption and nothing else around them, and to be screenshotted at the end.
What Is a Context Window? Why Long Chats Start Forgetting
A long conversation with an AI starts strong and then quietly stops holding on to what you agreed. People read that as the model losing the thread. It is simpler than that.…
The rest of the sets
Every storyboard
What Is an AI Agent? The Difference Between Answering and Doing
What Is MCP? The Universal Adapter for AI Tools
Every AI assistant that can touch your real work has the same problem underneath: your calendar, your files, and your notes all speak different languages. MCP is the agreement that fixed it.…
What Is RAG? Letting AI Take the Open-Book Exam
A plain-language visual explainer on retrieval augmented generation: why models cannot know your documents, how the right pages get found and handed over, and why the answer is only as good as the library.
What Are Tokens? How AI Actually Reads Your Words
Before an AI model can do anything with your message, it has to chop it up. Not into letters, and not quite into words either. It uses pieces called tokens, and once…
What Is an LLM? The Prediction Machine Behind the Chat
An LLM can feel like a person on the other side of the screen, but its core job is more mechanical and more interesting: it predicts what piece of language should come…
Research. Plan. Execute. Test.
A four-step way of working with AI: research the ground, plan the move, execute it, test what came back. Drawn step by step, with the failure modes marked. Based on the article…
Markdown: The File Your AI Tools Speak
What a Markdown file is, why so many AI tools read and write it, and how the same plain-text habit organises notes, prompts, and whole projects. What Markdown is, why it matters,…
Ollama + n8n: Local AI Setup
A step-by-step tutorial for setting up Ollama (local AI models) and n8n (workflow automation), then connecting them to build automated AI workflows on your own machine.
API Flow: Digital Conversations
A beginner-friendly explanation of how APIs work using a restaurant analogy. The client is a customer, the API is the waiter, and the backend is the chef.
LLM Thinking: Human vs AI
A visual comparison of how humans think sequentially (vertical, one path at a time) versus how LLMs think in parallel (horizontal, evaluating all possibilities simultaneously).
Cookie Monster: Privacy & SweetNet
An interactive storyboard about browser cookies, data tracking, and digital privacy. What actually happens when you click Accept All, drawn scene by scene.
What a Neural Canvas is
A Neural Canvas is a nine-panel illustrated explainer. Each one takes a single idea from artificial intelligence, such as how a language model reads your prompt or where your data goes when you accept a cookie, and draws it as a short story with ordinary objects.
You do not need a technical background. Read the panels in order, like a comic strip. The last panel is made to be saved or shared. Every canvas is free, and when you want the detail, each one points to a longer written entry in The Human Algorithm.