Nobody's job will look the same in a year, including ours. What a designer, an engineer or a strategist does changes with every project, so what we teach changes with it. These six courses are where that lands. The first is for everyone; the rest depend on what you want to make.
Learning
The work is changing faster than any course can keep up, so we made our own.
What everyone at Made by Many knows about AI, whatever their discipline. The parts, the tools, the rules we work to, and enough of how AI features are built to talk about them with a client.
11 lessons
A model, a harness and an agent are three different things, and what separates them is what each one can see and do.
Each job has a right tool, and three are approved for studio work: Claude, Granola, and Gemini in Google Workspace.
A handful of rules apply to every AI tool we use, because we answer to clients and to the people in our material for what happens to it.
Most bad output comes from a bad brief or an unchecked answer, so brief it like a colleague, work the conversation, and check anything factual against a source.
A connector gives an agent one of your tools with a sign-in rather than a build, and everything the agent can reach is something it can get wrong.
A conversation is private and forgets, so shared work lives in a shared place: a project or a folder, with a short document at the top that people and agents both read.
A skill is a way of working an agent can follow, and most of the time the right move is to use one that exists rather than write a new one.
Most AI features are the same few parts in different arrangements: a model call, the data it is given, the tools it may use, the guardrails around it, and a person where it matters.
An AI feature costs money to build, to run on every use and to keep working as models change, and a scope that settles the data and what good enough means before the build is the one that holds.
What the studio thinks about AI is published as principles and positions, so in front of a client you quote them, with their status, rather than improvise.
Every term the AI fluency course calls out, defined in a sentence or two, with the lesson that teaches it.
Set up your machine and learn the parts every project is made of: version control, how a frontend is put together, what a design system looks like in code, and how it gets onto the internet.
5 lessons
Claude Code and the tools the skills drive are installed once, the skills arrive with your studio account, and when something breaks the fix is a good description handed to the agent, not a command you have to know.
Git is checkpoints, and a pull request is how a change is proposed, reviewed and merged: every change is recorded, undo is always available, and nothing lands without someone looking.
A frontend is pages made of components, built on packages other people wrote. Most of what you know from Figma already has a name in code, and the fastest way to change something is to point at it.
Colour, spacing and type come from tokens rather than typed-in values, which is what a design system is once it is code.
A merge becomes a build becomes an address anyone can open, and every change gets a preview address before it reaches the real one.
Design and build a real thing with an agent: brief it, set up the design system, explore the options, and make the components and screens.
10 lessons
A real project, scaffolded by the suite and running in your browser inside an hour.
A design prompt says what the thing is for, who it is for, and what it must not be.
From a brand colour to a full token set, so everything made after this is on-brand by construction.
Refusing the first plausible direction is the skill: ask for several distinct options and name the axis each should vary on.
One component, with its three states, real content and a visible focus ring.
Compose components into a screen, then a second screen that shares them.
Taste stays in charge: a generated screen is a starting point you direct until it is yours.
A model flatters by default, so critique has to be asked for in a way that cannot be.
Contrast lives in the tokens and the keyboard pass runs on the working prototype, so accessibility is checked during rather than after.
What we generate, what we do not, and how every asset records where it came from.
Take the thing you made and make it work in the world: deploy it, give it memory, connect it to real tools, put a model inside it, and own it in production.
8 lessons
A push becomes a build becomes a URL someone else can open.
A page that has to remember something needs somewhere to put it, and that is what a database is.
A CMS or an API gives the thing real content, and the key that unlocks it never goes in the code.
A model behind a feature of the thing itself, reached through the Vercel AI Gateway.
Scheduled jobs and serverless functions do work while nobody is watching, and nobody looks after a server.
The audits say what is wrong with the thing before a person does.
Nobody hands it over: the person who made it promotes it and owns what is live.
A skill is a way of working written down where the machine can follow it, and you now have one.
The practical guide to turning an intent into a spec with an agent, and keeping that spec alive while the build corrects it.
2 lessons
An intent becomes a spec covering the problem, the scope, the riskiest assumptions and the success criteria.
A planning session is a working conversation that ends with a drafted spec, not a meeting that ends with notes.
The practical guide to using AI in research without mistaking a plausible account for something anyone said.
4 lessons
A concept in a deck earns an opinion about a deck. With agents, the thing itself is cheap enough to put in front of people instead.
Which of the approved tools does what in research, and what never goes near any of them.
When AI synthesis is safe, across sessions you were in with every finding linked back to the moment it was said, and when it is not.
A synthesis is only as good as the findings you can trace. This is the habit that catches the one nobody said.