Skip to main content
MessyDevAI is messy. Clean it up.MessyDev

The whole course

8 learning chapters, 31 lessons

Take a lesson, practise it, build something real, then ship it. Work through it in any order — Messy keeps track of what you've finished.

Messy the MessyDev robot mascot working on a laptop
Lessons teach the idea. The practice spaces are where it sticks. Do one of each this week and you'll feel the difference.

Build → Understand → Test → Fix → Learn → Launch

Chapter 1 — AI Foundations

How AI actually behaves, and how to get useful work out of it. 7 lessons, Lesson 1 free.

  1. Lesson 1

    Prompt

    Ask better, get better.

    3 exercises→

  2. Lesson 2

    Context

    What you feed it is what you get.

    3 exercises→

  3. Lesson 3

    API

    Wires, keys and status codes.

    3 exercises→

  4. Lesson 4

    Agent

    When AI does, not just says.

    2 exercises→

  5. Lesson 5

    Judgment

    Trust, but verify.

    2 exercises→

  6. Lesson 6

    Shipping

    Get it out of the garage.

    2 exercises→

  7. Lesson 7

    Lovable

    Brief it, iterate, ship it.

    5 exercises→

Chapter 2 — AI at Work

Use AI on the job without embarrassing yourself. 4 lessons, part of Pro.

  1. Lesson 8

    Meeting Notes

    Summaries someone else can act on.

    2 exercises→

  2. Lesson 9

    Voice

    Sound like you, not the model.

    2 exercises→

  3. Lesson 10

    Numbers

    Check it before you forward it.

    2 exercises→

  4. Lesson 11

    What Not to Paste

    Some things don't go in the box.

    2 exercises→

Chapter 3 — Vibe Coding

Describe it well, and the build keeps up with you. 3 lessons, part of Pro.

  1. Lesson 12

    Brief

    Describe it well and it builds itself.

    2 exercises→

  2. Lesson 13

    Iteration

    One change at a time, verified.

    2 exercises→

  3. Lesson 14

    Read the Build

    Understand the thing you shipped.

    2 exercises→

Chapter 4 — Builder Lab

Databases, auth, payments, email — the real plumbing. 4 lessons, part of Pro.

  1. Lesson 19

    Data

    Shape the tables before the app shapes them for you.

    3 exercises→

  2. Lesson 20

    Access

    Decide who can see and change what — before someone else does.

    3 exercises→

  3. Lesson 21

    Checkout

    Take money without holding the card.

    2 exercises→

  4. Lesson 22

    Wire

    Connect the pieces: email, webhooks, quiet failures.

    4 exercises→

Chapter 5 — Agent Lab

AI that does things — tools, function calling and MCP. 4 lessons, part of Pro.

  1. Lesson 15

    Function

    Teach a model to call your code.

    3 exercises→

  2. Lesson 16

    SDK

    Talk to the model from your own code.

    3 exercises→

  3. Lesson 17

    MCP

    Plug agents into real systems, safely.

    3 exercises→

  4. Lesson 18

    Eval

    Prove it works before you trust it.

    3 exercises→

Chapter 6 — Automation

Workflows, triggers and webhooks that run without you. 3 lessons, part of Pro.

  1. Lesson 23

    Trigger

    Every automation starts with something a machine can notice.

    2 exercises→

  2. Lesson 24

    Hook

    Let other systems poke yours — without letting anyone poke it.

    2 exercises→

  3. Lesson 25

    Recovery

    Green runs that did nothing, and how to make them admit it.

    2 exercises→

Chapter 7 — AI Safety & Trust

Secrets, hallucinations, privacy and the stuff that bites. 3 lessons, part of Pro.

  1. Lesson 26

    Secrets

    Keys, where they belong, and what to do when one gets out.

    3 exercises→

  2. Lesson 27

    Injection

    When the document you fed the model starts giving it orders.

    3 exercises→

  3. Lesson 28

    Honesty

    Confident sentences, no evidence — and what to share when asking for help.

    3 exercises→

Chapter 8 — Career Paths

Turn what you can do into something other people will pay for. 3 lessons, part of Pro.

  1. Lesson 29

    Evidence

    Adjectives out, artefacts in.

    2 exercises→

  2. Lesson 30

    Roles

    Where this skill actually gets paid.

    2 exercises→

  3. Lesson 31

    Pitch

    Asking for the time, the budget or the job.

    2 exercises→

Where you practise

Practice

Repeat the basics until they feel obvious.

Build

Turn an idea into a real thing with Messy.

Diagnose & fix

Something is broken. Read it, find it, fix it.

Ship

Practice shipping without breaking anything.

Tool Lab

Try the tools before you trust them.

Your evidence

Proof of what you can actually do.

Mentors

Real people when Messy isn't enough.

All eight chapters, open now

Every chapter is live and every lesson is hands-on. Here is what each one teaches.

Chapter 1Open

AI Foundations

How AI actually behaves, and how to get useful work out of it.

    Chapter 2Open

    AI at Work

    Use AI on the job without embarrassing yourself.

    • Meeting notes that hold up — Summaries someone else can act on.
    • Writing with a house voice — Keep the tone yours, not the model's.
    • Data you can defend — Checking numbers before you forward them.
    • What not to paste — Confidential material and safer alternatives.
    Chapter 3Open

    Vibe Coding

    Describe it well, and the build keeps up with you.

    • Briefs that build — Scope, screens, data, done.
    • Iterating without thrash — One change at a time, verified.
    • Reading what it made — Understanding the thing you shipped.
    Chapter 4Open

    Builder Lab

    Databases, auth, payments, email — the real plumbing.

    • Tables and relationships — Modelling the thing you're building.
    • Logins and permissions — Who can see what, and why.
    • Taking money safely — Checkout, subscriptions, webhooks.
    • The wiring between — Email, webhooks and quiet failures.
    Chapter 5Open

    Agent Lab

    AI that does things — tools, function calling and MCP.

    • Function calling — Tool schemas a model can actually use.
    • SDKs and keys — Calling models from your own code, safely.
    • MCP servers — Connecting agents to real systems.
    • Evals and harnesses — Proving it works before you trust it.
    Chapter 6Open

    Automation

    Workflows, triggers and webhooks that run without you.

    • Triggers and actions — The shape of every automation.
    • Webhooks — Letting other systems poke yours.
    • Failure handling — Retries, alerts and quiet breakages.
    Chapter 7Open

    AI Safety & Trust

    Secrets, hallucinations, privacy and the stuff that bites.

    • Secrets and keys — Where they go and where they never go.
    • Prompt injection — When your input isn't your input.
    • Being honest about limits — Saying 'I checked' and meaning it.
    Chapter 8Open

    Career Paths

    Turn what you can do into something other people will pay for.

    • Evidence over claims — A portfolio built from real work.
    • Role-shaped practice — Marketer, ops, founder, developer.
    • Talking about AI work — Interviews and internal pitches.

    Start small today

    One short exercise a day is enough to build the habit.

    Open today's exercise