Course catalog
Build with ChatGPT and Codex: in the apps first, then the cloud, then the command line.
OpenAI gives you more than one place to build: a chat window, a side-by-side Canvas editor, packaged custom GPTs, Codex tasks that run in their own cloud container, and a Codex agent on your own machine. Each one is good at a different job, and picking wrong costs you an afternoon. This course walks the real surfaces, starting in the apps — the model picker labels, Canvas revisions, the custom GPT configuration view — then cloud tasks and code review on a pull request, with the command line as a power-up: the install and login, the sandbox and approval settings, config.toml, and AGENTS.md. You finish by comparing app, IDE, cloud, and terminal honestly, moving one task between them without re-explaining it, and deciding when a subscription seat beats paying per token.
Section 1
Choose the right ChatGPT mode and entry point, edit durable artifacts in Canvas or blocks, set reasoning deliberately, package reusable work, and review Codex's changes in the desktop app before you keep them.
Section 2
Delegate bounded repository work to Codex Cloud, let Codex prepare and publish the environment, review evidence against the right baseline, and carry the result through a pull request.
7-day free trial on every plan for first-time members · Cancel anytime
Get startedSection 3
Describe a small hosted web app in ChatGPT or Codex, check it as a visitor, save a version before you deploy, and decide who can visit, edit, and sign in. Sites is in public beta, and availability depends on your plan, region, and workspace.
Section 4
Add the terminal to what you already do in the apps: launch Codex in an explicit working directory, then set project instructions, configuration, the sandbox boundary, the approval policy, and a verification loop.
Section 5
Choose among Chat, Work, Codex, IDE, terminal, cloud, Sites, plugins, and automations by task shape; hand off state and separate subscription seats from API metering.