Build and Ship an AI App
Build software with AI agents, and build AI features into software. Add an assistant, answers from documents, evals, an MCP server and defences against prompt injection to a real Django app, then build and deploy a web app of your own.
๐ป Some quests are built on your computer, with Python and the free neugen tool. The course walks you through setting it up.
Already have an account? Log in and you'll come back here.
About this course
You'll learn two things side by side: how to build software with AI agents (Claude Code, Codex, Gemini CLI and others), and how to build AI features into software.
Most quests give you a working Django app and ask you to add one feature: an AI assistant that uses the app's data through tools, answers from the organisation's documents, an eval suite, an MCP server so people can use the app from Claude or Codex, and defences against prompt injection. You'll use a coding agent throughout, and learn to brief it well and check its work.
It ends with a capstone: your own AI web app, planned in the first quest, built from scratch, and deployed on the internet for free. (Free hosting doesn't cover model calls: budget a few dollars for your own API key.)
You'll need: comfortable Python, the command line and basic git. Some HTML and HTTP helps. Any laptop works.
Stack: Django, server-rendered templates with plain JavaScript and CSS, SQLite (PostgreSQL for your capstone).
Decide what you'll build
Write a short, provisional product brief (about 500 to 800 words) for the AI web app you will build and deploy in the capstone: who it is for, one core workflow, the data and documents, what good answers look like, the risks, and what you will not build. It must fit the course's product envelope.
Plan your app
Decide what you'll build and ship at the end of the course.
Work with a coding agent
Build a small app with an agent, then add a real feature to an app someone else wrote.
-
Working with agents ๐ Locked
Step 2: Pair with an agent: build a small Django app
Your first project with a coding agent. Install a harness, write a project instruction file, and build a small Django app with logins from a fixed spec: plan first, work in small checked steps, read every diff and commit checkpoints, and explain how you checked the agent's work.
Unlocks after step 1.
-
Working with agents ๐ Locked
Step 3: Add a feature to an existing app: a room calendar
Real work is mostly adding features to code someone else wrote. Explore the course app with a coding agent until you understand it, then add a week calendar for each room: the right week, clear statuses, and privacy for members. Build it in small, checked steps, then write tests for what you built and review the tests your agent wrote.
Unlocks after step 2.
Add AI to the course app
Give the app an AI assistant that answers from its data and documents, then open it to Claude and Codex.
-
AI features ๐ Locked
Step 4: An AI assistant inside your app
Add an AI assistant to the course app with the official openai SDK. It answers questions from the app's data and from the handbook (with the file it cites), always with the logged-in user's permissions, and it can make one kind of change, but only after the user presses Confirm. Write the tool-calling loop yourself, keep your API key out of the code, and test the assistant with a fake model and a few real questions.
Unlocks after step 3.
-
AI features ๐ Locked
Step 5: Add MCP support
Open the course app to Claude and Codex. Write an MCP server by hand in Django: the Streamable HTTP transport at /mcp, JSON-RPC, personal tokens that are stored hashed, and the assistant's tools with clear names, tight schemas, read-only hints and helpful errors. Every tool runs as the token's user, with exactly the web app's permissions, and changes still wait for the user's Confirm.
Unlocks after step 4.
Capstone: your own AI app
Build the app you planned, ship it on the internet, and show it.

