You have seen the post. Someone ships three screens and a demo video over a Saturday, captions it "I built this in a weekend," and the internet rewards it with reposts. This flex should be retired, because it celebrates the one thing about building that has stopped being hard and stays silent on the only thing that was ever hard.
Speed to demo is the cheapest metric there is
Getting to a first working screen used to be a real signal. It meant you had the skill, the focus, and the hours to make something run. It is not that signal anymore. With agents, speed to first output is close to free, which is the whole point of vibe coding. The weekend build is vibe coding's highlight reel.
So the flex now measures the cheapest, least durable thing you can produce. It tells you the demo runs. It tells you nothing about whether the thing survives a real user, a real edge case, or its own second month. You are showing off the part of the cost curve that comes before maintenance starts to bite, and presenting it as if it were the whole graph. Speed to demo is time to the cheapest, most brittle version of the thing, and time to durable value is a different number that has almost nothing to do with it.
Building fast is fine, the flex is the problem
Building something in a weekend is a great skill and a great way to learn. A prototype you build to answer a question and then throw away is healthy, and it is most of what instant prototyping is for. The problem is not the speed. It is treating the speed as the achievement.
The flex broadcasts time to first output as if it were proof of a durable product, and then, far too often, the weekend build does not get thrown away. It gets a demo, the demo gets a customer, and it quietly graduates into production without anyone ever adding structure. The build that was cheap to make becomes the thing that is expensive to own. A healthy prototype gets thrown away after it answers its question. A vanity build gets left running because tearing it down is awkward. Watch what happens after the demo.
The only question worth asking
Next time you see "I built this in a weekend," do not ask how. Ask whether it was still running the weekend after. Ask if it has an eval that defines done. Ask if it survives a regression, or if it would still run with no one maintaining it. Those questions measure time to durable value, which is the number that decides whether you built past the crossover point or stopped on the brittle side of it. I built Falkster to a real revenue number as a company of one, and the reason was never weekend speed. It was that the thing keeps running when I am not touching it. That is the flex worth having, and it does not fit in a Saturday.
Find the most impressive demo your team shipped this quarter, the one that got the reaction. Ask the weekend-after question about it honestly. If it would not survive a month unattended, you do not have a win, you have a vanity build accruing debt. Throw it away on purpose or put structure under it on purpose, but stop letting it sit in the middle collecting applause and interest at the same time.