Vibe Coding

In this course you'll learn to build complete applications using AI tools such as Claude Code. No prior programming background required. We cover the whole journey, from a first rough idea to a working application running in production.

The course runs in three stages:

• Getting to grips with AI and vibe coding tools. How LLMs actually work, what tools are out there, and which one to reach for when.
• From idea to working product. You build your own application step by step: prompting effectively, specifying features clearly, and working in a professional development environment. Along the way you'll pick up the software development fundamentals needed to put together something robust.
• Working professionally and shipping. You sharpen your workflow with tests, type checking, version control and automated feedback loops, then finish by deploying your application to production.

In short, you cover the full spectrum of modern AI-driven software development. By the end you'll be able to turn your own ideas into working applications independently, and put AI to work as a real development tool.

Day 1: AI foundations, tooling, and your first application

• What's going on under the hood: how AI models generate output, why tokens and prompts matter, and how tool calling works in practice.
• The tooling landscape: a tour of what's available, Claude Code among others, and where these tools overlap and where they diverge.
• Prompting that works: expressing what you want precisely enough that a coding agent picks it up without misfires and translates it into code that runs.
• Specifying features: describing functionality from the end user's point of view, anchored in business value, so you reach a working result quickly.
• Instruction files: using files such as CLAUDE.md or AGENTS.md to keep your coding assistant supplied with the right context every time.

Day 2: Building properly, safeguarding quality, and going live

• Local development with AI coding tools (Claude Code, Codex and similar): finding your way around a project's folder structure, and starting and testing an application from the terminal.
• Version control: applying Git so changes to your codebase stay under control.
• Architecture and engineering basics: an introduction to CRUD operations on data and how databases work.
• Automated feedback loops: setting up tests your coding agent can run on its own to verify the application still behaves correctly.
• Going live: closing out by deploying your application to a public URL that anyone can reach.