A post-2000s kid programmed with AI for 10 days. CHEN Tianqiao invested 30 million in 24 hours. He became the CEO with his senior-year assignment.
A senior undergraduate student spent ten days on Vibe Coding and secured 30 million yuan in investment in just three months, turning his graduation project into a company! As the domestic open - source project MiroFish topped the GitHub trending list, the era of "super individuals" where one person can create a global hit has truly arrived.
Waking up one day, the domestic open - source AI project MiroFish topped the GitHub trending list!
Surprisingly, MiroFish, which outperformed open - source projects from institutions like OpenAI, Google, and Microsoft, was developed by a domestic senior undergraduate student. In just three months, he secured 30 million yuan in investment.
By relying on an AI open - source project, did he directly turn his graduation project into a startup and become the company's CEO?
This is not a myth but a real story that happened in China in the AI era.
Even more amazingly, his previous project, BettaFish, also topped the global GitHub trending list.
Both of these projects reached the top of the list, and they were both developed by this post - 2000s kid through Vibe Coding in just ten days.
Link: https://mp.weixin.qq.com/s/UyYVjlBCvQRJI6B_MmZbsA?scene=1&click_id=22
When we look into the details, we'll find the underlying significance of this era: The era of super individuals has truly arrived!
When the power of AI tools, open - source communities, and young developers combines, one person can be an innovation army.
In the future, there will be more super individuals creating more startup myths in the AI era.
A senior undergraduate student's three months: From going viral on GitHub to securing 30 million yuan in investment
The story began quite ordinarily.
In the summer of 2025, senior undergraduate student BaiFu was still anxious about internships.
At that time, his future educational path was settled, so he decided to finish his graduation project in advance so that he could focus on finding an internship.
Instead of writing tens of thousands of lines of code from scratch, he opened tools like Claude Code and started the then - popular development method - Vibe Coding.
Within ten days, he integrated his years of university accumulation into an open - source project called BettaFish.
This is an AI public opinion analysis tool that can automatically collect online public opinion information and generate structured analysis reports.
"Micro - Opinion" helps people break through the information cocoon, restore the original appearance of public opinion, predict future trends, and assist in decision - making
Users only need to submit analysis requests in a chat - like manner, and the intelligent agent will start fully automatic analysis of more than 30 mainstream social media platforms at home and abroad and millions of public comments.
Technically, it combines automated data collection, large - model analysis, and automatic report generation, and can finally output a public opinion analysis document of dozens of pages.
For many enterprises, such work usually requires an entire team, while BettaFish attempts to automate this process with AI.
However, when BettaFish was just completed, it wasn't given much hope. Even BaiFu's doctoral senior criticized it directly after seeing it: "This is not a qualified graduation project. You need to spend more time revising it."
After reaching 1000 stars, a miracle occurred
After the project was launched, there wasn't much of a stir.
Like many independent developers, BaiFu tried various forms of promotion, such as submitting articles to open - source weeklies and official accounts, posting videos on Bilibili, and posting on technical communities, but the results were not obvious.
After hard promotion, the project finally reached 1k stars. Unexpectedly, this became the first traffic turning point for BettaFish.
Regarding this, BaiFu was already very satisfied, so he wrote a summary article and posted it on Xiaohongshu, titled "What can an open - source project with over 1K stars bring to a college student?"
Who could have imagined that a miracle would occur a week after the article was published!
The article was reposted and recommended by several big V - users in the technical community.
Subsequently, the GitHub page of BettaFish began to experience a significant traffic explosion, and the number of stars increased at an astonishing speed - 5K, 10K, 20K, 30K...
In just a few days, the project gained more than 20,000 new stars!
As a result, BettaFish quickly topped the global GitHub trending list and became one of the most discussed open - source projects in the domestic AI development circle at that time.
After the project became popular, a large number of emails began to flood into BaiFu's inbox, such as internship invitations from big companies, technical cooperation invitations, and even olive branches from investment institutions.
After that, the number of users of the project increased, and the related demands exploded! The surge in a large number of users brought him great pressure.
The significant phone call
After hesitating and being in a state of confusion for a long time, Baifu decided to go all - in on entrepreneurship. At this time, a phone call changed the course of events.
Chen Tianqiao, the founder of Shanda Group, showed great interest after seeing BettaFish and hoped to have further communication with him.
At that time, BaiFu's inbox was full, and he hardly had time to reply. However, the Shanda team kept contacting him persistently. Finally, he was touched by their enthusiasm and agreed to have a chat.
During the subsequent conversation, BaiFu was impressed. He truly felt that Shanda was more serious about "embracing super individuals" than any other company.
Chen Tianqiao said bluntly that the technical level of BettaFish is not particularly outstanding.
However, what Shanda valued was BaiFu's complete planning of the entire process from data collection, analysis to prediction in the project, as well as his ability to "identify and define real and valuable problems and try to solve them in new AI - based ways."
After hanging up the phone, BaiFu didn't hesitate at all. He packed his bags and went to Shanghai alone to start cooperation with Shanda.
The birth of MiroFish: Developing the second project in ten days
With the support and full freedom given by Shanda, BaiFu began to implement a new idea.
If BettaFish solves the problem of analyzing the past, then the new project aims to predict the future.
The core idea of this project named MiroFish is to build a digital sand table that simulates the real world.
In essence, MiroFish is an AI prediction engine driven by multi - agent technology.
It can extract seed information from the real world, such as breaking news, policy drafts, and financial signals, and automatically build a high - fidelity parallel digital world.
Within this space, there are thousands of intelligent agents, creating a "parallel universe."
Each intelligent agent has an independent personality, long - term memory, and behavioral logic, and can interact freely and undergo social evolution.
This time, BaiFu chose Vibe Coding again and was still incredibly fast -
The development of MiroFish was completed in just ten days.
That night, looking at the working program, BaiFu was so excited that he recorded a rough demo and directly submitted it to Chen Tianqiao.
Within 24 hours of submitting the video, the two had a second conversation, and Chen Tianqiao immediately decided: The group will invest 30 million yuan to fully support the in - depth incubation of MiroFish!
Overnight, BaiFu changed from an intern at Shanda to the CEO of an AI startup.
MiroFish has gone further than BettaFish:
The end - point of BettaFish's analysis becomes the starting - point of MiroFish's prediction.
From raw data to intelligent decision - making, a real closed - loop begins to form.
The goal of the MiroFish team is clear: to build an AI prediction system that can simulate the real world. From financial markets to social events, from business decisions to creative experiments, it allows the future to evolve in the digital world in advance.
Any prediction task that can be monetized in reality is the target scenario for MiroFish.
Target application scenarios of MiroFish: financial decision - making support, policy and public opinion prediction, public relations crisis simulation, marketing strategy testing, story and fiction deduction, academic research support
Currently, the project has 5.7K stars on Github🌟. The experience website is as follows👇:
Link: https://666ghj.github.io/mirofish-demo/
Two consecutive open - source projects both topped the GitHub trending list, and both were completed in ten days.
BaiFu has successfully proven that "Vibe coding has infinite potential," and he himself is a Chinese example of a super individual