Build software with AIplanned
You've been promoted. Your agents are your employees.
This is the setup behind everything else on this site, and behind the production software I ship at work. How the folders are laid out, what the rules files say, which guardrails are non-negotiable, and how to give your agents access to the tools you already work in.
You'll need a paid Claude plan, which is $20 a month, and a machine you can install things on.
What It Covers
Running agents the way you'd run a team.
You finish with a configured machine and a setup that works the way you do.
Setting up Claude Code
Account, plan, install, and the choice between working in WSL or in your IDE. Ends with the setup repo installed, which brings every plugin, command and tool in this course with it, plus the usage bar so you can see what you're spending.
How the money works
Tokens, context windows, model choice, and what each of them actually costs you. Most people either burn budget without noticing or pay for a bigger model on work that never needed one. Both are avoidable once you can see what's happening.
Your dev folder
One folder at the top with every repo sitting inside it, and Claude opened at that level so it can see all of them at once. This is the change that lets you work across several projects in a session, and almost nobody does it.
A codebase somebody else wrote
Everything so far assumes you started the project. Most work is the opposite: an old repo, no rules file, and conventions that only exist in the code itself. How to get Claude oriented in it, and how to make it follow patterns nobody ever wrote down.
CLAUDE.md as an onboarding doc
You're a team lead now. This file is how you brief your employees on their first day: how you work, how tests get written, what the CI does, what nobody is allowed to touch. Covers the hierarchy too, since the dev folder, a project and a repo each want a different file.
Making it write like you
Every model writes the same way and people can feel it. The fix is the same artifact as the last lesson, pointed at prose. Banned constructions, the cadence you actually use, the words you would never say out loud, and a checklist to run before anything goes out. Then it drafts and you edit.
Guardrails that actually enforce
Written rules are a request. Permissions and hooks are a wall. Which guardrails are nice to have and which ones you genuinely can't ship without, then how to write a hook that fails closed so a bad command never runs in the first place.
Slash commands and skills
Anything you do more than twice becomes one keystroke. Pre-PR review, vulnerability fixes, whatever your work actually repeats. Every command in the setup repo, what it does, and how to write your own.
Agents
When handing work to a subagent is worth it and when it quietly costs you money for nothing. How to scope what you delegate, and why the answer that comes back needs the same scrutiny as anything else.
MCP
Wiring Claude into the tools you already use, so it can read your tickets and act on them directly. This is the biggest multiplier in the course and most people never turn it on.
Know when it opens.
Leave your email and we'll tell you when the first lessons are up. You'll hear from us once, at the point where there's actually something to watch.
Everyone on the list pays the founding price of $149 a course. It opens at $199. Nothing to pay now, and no date attached to any of it.
Would rather we just built it for you? See what that costs.