The learning map
The ideas worth knowing about AI, laid out in five stages, from your first chats to building your own tools. Rucksack plugs into the AI app you already use, such as ChatGPT or Claude, and turns it into a guide that teaches you each idea when it's useful.
Essentials to chat confidently with AI.
- How AI worksChatGPT, Claude, Gemini and Copilot are AI apps: what you open and type into.
- Context: what your AI seesIn a new chat, the model reads only what the app puts in front of it: a few things, usually these three.
- Custom instructionsYour own instructions are a setting in your AI app where you write lines of your own, which the app puts in front of the model before your first message in every chat.
- Memory in AI appsThe app's notes about you are short facts your AI app saves between chats, often without being asked.
- What not to shareA message you send in an AI app goes to the company that runs it.
- Writing a good promptAn ask is what you type to your AI app to get something done.
Learning how to get more out of AI.
Customising how AI works for you.
Building your own AI tools and skills.
Expanding the reach and scale of your AI activities.
Learn by building
Take on any project you like, from a plan for work to a small app of your own, and your AI will help you work out the steps and get it done. You learn what you need along the way.
Or start with one of the projects below. Each one is a practical way to learn what matters about AI while you build something you'll keep using: an AI app where every new chat already knows you and what you use AI for, an automation that watches a topic, price or page you care about and writes you a short report with sources, or a web page that shows an employer or client what you can do with AI.
-
Trailhead
Your DaypackRucksack in the AI app you already use.
A few lines in its settings and a saved note in its memory, so every chat knows how you like to learn, what you use AI for and what never goes into a chat. A chat named "Daypack" is where you learn on purpose.
AI and meA short guided conversation with your own AI about the big picture: how you feel about AI, what excites you, what worries you, where you want it in your life and where you don't. -
Day walk
A week with your DaypackA week of your real work in your AI app, with a handful of habits tried once each. -
Base camp
Your RucksackThe plain text files your AI apps read first: who you are, how your AI should work with you, your apps, what's worth remembering, your skills, your projects, your journal.
They're yours, in a folder on your computer or your own GitHub.
-
High camp
The scoutYour first automation, built with your AI: a job that runs on a schedule while you get on with your day, and only ever writes you a report.
On the way, you learn the kinds of automation and how to choose one.
-
Summit
Your portfolioA web page you build and put online with your AI, showing what you can do with AI.
Other ideas
From Base camp up, whenever you like. Each can sit in your portfolio.
- Your own connector: so your AI can reach a tool you use.
- An app or game: one you build by describing it.
- Your own model comparison: AI models on the same task, scored by you.
- A team of agents: several AI agents working on one job.
How the learning works
The map above shows everything there is to learn: the ideas worth knowing about AI, grouped under five waypoints, from Trailhead at the start to Summit at the end. Each idea is a concept, such as why AI can be wrong and still sound sure, or how to word a request so the answer fits. Every concept comes with a plain explanation, tips and something to try. The map is the same for everyone, and this page doesn't track you.
You don't study the map. The learning happens in the AI chat app you already use, once you've set up Rucksack there. Rucksack is a free set of instructions and files that turn your AI app into a learning guide. You can set it up as a Daypack, a few lines pasted into the app's settings, or as a full Rucksack, a folder of files that ChatGPT and Claude can connect to and read. When something comes up in your own chats, your AI gives you a short tip that fits what you're doing, and when you ask it to teach you something, it goes further. It pitches the tip or the lesson to how you like to learn and what you use AI for, which it asks you when you set it up.
A full Rucksack also keeps your progress. When you use an idea on real work or explain it back in your own words, your AI records that you've got it, and a page in your Rucksack shows your progress, with that concept's circle filled in. Learn more than half of a waypoint's concepts and you've completed that waypoint. There are no tests or scores. A Daypack keeps no progress record.