Now you turn the brief into a running app. This lesson is where "I directed an AI to build software" stops being an abstraction and becomes a thing you watched happen.

Paste your brief into Lovable and it goes to work. Since mid-2025 Lovable runs in an agent mode: it does not just spit out one response, it breaks your request into pieces, builds each, checks its own work, and shows you a running preview. You will see a real app appear on the right side of the screen in under a minute. That first moment (a working thing, from a sentence) is the one everyone remembers.

Then the actual skill begins, which is iterating. Your first version will be roughly right and specifically wrong in a few places, and that is normal and expected. You improve it two ways. You prompt for changes in plain language ("make the submit button larger and move the results above the form"), and you edit visually, clicking directly on elements to nudge layout and style without describing them. You will move between these constantly. The rhythm is: look at what it built, decide what is wrong, make one change, look again. This tight loop (small change, review, repeat) is how every good build proceeds, and it is a habit worth forming now because it carries through every tool in the course.

One thing to internalise while you build: when the output is not what you wanted, resist the urge to blame the tool and instead reread your prompt as if you were the literal-minded junior receiving it. Nine times out of ten the fix is in the brief, not the tool. This is the same lesson as the previous section, now felt in your hands rather than heard in the abstract.

By the end of this lesson you will have your capstone's core screen and core action, built and running in a preview, ready to get its memory and its login in the next lesson.