MANSAFrançaisSign in

MANSA

Course 1: Building with AI, as a developer

Students, young graduates and early-career developers who want to work with an AI assistant, not just try it.

Sign in to start
The tree of an AI-assisted project: .claude/skills, pm/tasks and src/ branching from the project root

What you will be able to do

  • A working local development environment, with an AI coding assistant installed and pointed at a real project
  • A clear mental model of how an AI-assisted project is organised — three schemas you will reuse on every project
  • Built a small but real AI-driven web application yourself, with your assistant, from an empty starter to a working app

Modules

First contact

Not started

Before any lesson, open a chat with the AI and ask it eight simple questions, one at a time. For newcomers this is essential; for experienced people, we still recommend it. The point is not to learn facts. It is to get comfortable: to see what the AI is, what it is not, and how it talks.

Start with First contact
01

Primer: converse, don't just prompt

Not unlocked

Understand the one mindset shift the whole course is built on, and the four ideas that make an AI assistant useful on real work — conversation, skills, a shared memory folder, and a habit of improving after every task.

  1. The intention
  2. Learn to converse, not just prompt
  3. Know its skills
  4. Know the pm/tasks folder
  5. The improvement loop
  6. The work cycle you'll use
This module isn't unlocked for you yet.
02

Environment setup

Not unlocked

Install the tools, sign in to your AI assistant, get the course project running on your machine, and point your assistant at it.

  1. What you need
  2. Install
  3. Sign in to your assistant
  4. Get the course project running
  5. Point your assistant at the project
This module isn't unlocked for you yet.
03

The three AI schemas

Not unlocked

Learn the three schemas that every AI-assisted project ends up needing once it goes past "one prompt in a chat window": where things live (the tree), how a change moves (the workflow), and how the system improves itself (the cycle).

  1. The tree — where things live
  2. The workflow — how a change moves
  3. The cycle — how the system improves itself
CheckpointAvailable once every lesson is done
This module isn't unlocked for you yet.
04

Build the demo app with your AI assistant

Not unlocked

Starting from the almost-empty starter, build a small AI-driven web app with your assistant, following the workflow from Module 02, phase by phase. You already ran the finished version in Module 01; now you make your own.

  1. The use case
  2. How it works
  3. Build it, phase by phase
  4. Compare with the reference build
This module isn't unlocked for you yet.
05

Homework & closing project

Locked

Extend your app on your own, then close the course the way every task should close: with a look back.

  1. Homework — pick one
  2. Closing project
  3. Annex — If you come from microservices