Lin Wei cried a little. Not from the answer, but from the absence of cruelty. No one mocked the question. No one demanded a "proof of effort." It was just help, given freely.
Revenue became a problem. Without VC money, they introduced a "Patron Pass"—a voluntary subscription for users who could afford it, which unlocked cosmetic tree skins and nothing else. To everyone’s surprise, 12% of users signed up within the first month. They weren’t paying for features. They were paying to keep the lights on. Today, StudyKaki is not a unicorn. It is not a household name. It has 2.3 million users—modest growth by tech standards—but an extraordinary retention rate: 78% of users who join stay for more than a year.
That was the seed. Lin Wei was not a coder by training—he was a mechanical engineering major—but he knew enough Python to scrape data and build a basic web interface. He called his creation StudyKaki (a play on study buddy and the Indonesian word kaki , meaning "foot," as in "on foot"—a journey taken together). studykaki
And the original noodle stall? There’s a small sticker on the cash register now. It reads: “Proud parents of StudyKaki’s founder.” His mother still doesn’t understand what a Laplace transform is. But she knows this: her son built a place where no one has to study alone.
He tried the usual solutions: YouTube tutorials (too passive), online forums (too toxic and competitive), and paid tutoring (too expensive). One night, at 2:00 AM, while trying to decipher a particularly vicious Laplace transform problem, he wrote in his notebook: “What if studying didn’t have to be a solo sport?” Lin Wei cried a little
They rolled back the AI tutor to a "last resort" button. They removed seed farming by capping daily reputation gains. They introduced a "slow lane" for the whiteboard—answers took at least one hour to appear, forcing users to think before typing.
Within a month, 200 users had joined. Within three months, that number grew to 2,000. By early 2020, StudyKaki had evolved. Lin Wei had dropped out of his master’s program (to his parents’ horror) and brought on two partners: Maya , a UX designer who hated how ugly learning platforms were, and Jun , a backend engineer who had been laid off from a failing fintech startup. No one demanded a "proof of effort
Worst of all, a new feature—an AI tutor that Lin Wei had reluctantly added to compete with ChatGPT—began answering questions instantly. And while it was efficient, something was lost. Users stopped explaining why an answer worked. They just pasted the AI’s output and moved on.