Entrepreneurs in the AI era are being redefined—An observation of the Xiaohongshu Hackathon Peak Competition

Author: Lawyer Shao Shiwei

From April 7 to 10, 2026, Xiaohongshu held a Hackathon Summit in Zhangjiang.

Two hundred contestants, working in a closed environment for forty-eight hours, building an AI software or hardware product from scratch.

This article will share my feelings on-site. It’s a message to all entrepreneurs in the AI era who are still observing or already on their way.

1. The lowering of technical barriers is a good thing

First, let me mention a striking data point from earlier. This year, Anthropic hosted the Claude Code Hackathon, with 13,000 registrants. Among the top five winners, only one had a design engineering background; the rest were lawyers, doctors, infrastructure workers, and professionals from other industries.

Five years ago, that would have been unimaginable. Back then, if you wanted to develop a software product, the first step was to recruit a technical partner because you couldn’t get past the coding hurdle. Now, AI has helped you clear most of that hurdle.

What does this mean? It means that the ticket for product innovation is no longer determined by whether you can write code, but by whether you understand the true needs of users.

And this is precisely the domain of cross-industry innovators. At this Xiaohongshu Hackathon Summit, I also saw lawyers participating—something that surprised and delighted me.

A lawyer knows better than an engineer what clients and their families need; a doctor understands what patients feel most anxious about during follow-up visits; a person who has been in a wheelchair for years knows better than any capable product manager what a disabled person truly needs. In the past, these insights were blocked by my own lack of programming skills. Now, the barriers are greatly lowered.

The influx of cross-industry talent is the real source of product innovation.

(Photos from the event)

2. What can be achieved in 48 hours? The instinct to start from pain points

The format of this Xiaohongshu competition is itself very interesting—48 hours of closed development. When time is pushed to the limit, people have no space for showmanship. You can only focus on the problem you know best, that hurts most, and that you most want to solve.

All the most touching projects on-site followed this rule.

The brain-controlled wheelchair won first place in the hardware track. The developer himself had experienced two severe paralysis episodes and is a wheelchair user. The device uses a hybrid of EEG and EMG signals to control the wheelchair, with the headset designed to look like a golden hoop. Someone trapped inside their body, personally building a golden hoop for themselves. This is a pain-driven product.

Using the “EEG + EMG” hybrid control technology, the user wears a headband, concentrates their thoughts, and makes tiny head muscle movements to control forward, backward, and turning motions of the wheelchair. To make operation accessible, he simplified the complex control logic, keeping only the core movements.

His wife, “Mushroom,” is the team’s “brainstormer.” Originally a web novel writer, she crossed over into a tech developer for this project. Two people, one wheelchair, one golden hoop, 48 hours. Bovine (the speaker) said on stage that he wanted to create more than just a product—he wanted to offer a new possibility for those trapped inside their bodies: to reconnect with the world using the faintest signals, supporting their gradual recovery.

Listening to this, I realized that all the most moving products in this competition are backed by developers who are also the users themselves.

There are also other interesting products.

Mind Kit addresses the problem of being interrupted during deep conversations with AI—breaking dialogue into independent thought nodes, allowing AI to process multiple threads in parallel, eventually forming a topological map of thoughts. The creator was fed up with this issue, so the product is sharp and impactful.

Another team, with an average age of 13, is made up of middle school students. They built a tool to help content creators craft viral Xiaohongshu posts. They are avid users of Xiaohongshu themselves, constantly thinking about how to grow followers, so their understanding of this scenario surpasses most adult professionals.

Feedback from the audience shows that popular products are not those starting from “what AI can do,” but those based on “what problems I (or those around me) are actually facing.”

The 48-hour format amplifies this pattern. It forces developers to skip all the “cool-sounding” pseudo-needs and focus on finding the real problems.

3. Xiaohongshu is the most suitable stage for this transformation

A special note on Xiaohongshu: its uniqueness isn’t in its traffic volume. Its uniqueness lies in integrating five activities—discovering pain points, open publishing, real-time feedback, precise search, and closed-loop conversion—that were previously scattered across GitHub, Twitter, search engines, and e-commerce platforms, into one domain.

In the past, product development path was: have an idea → seek funding → recruit people → develop → launch → find users. Each step had high barriers and required capital backing.

Now, the path is changing to: have an idea → AI-assisted development → post on Xiaohongshu → gather feedback → decide whether to scale.

The steps cut out are the most expensive ones of the past decade. This means the cost of trial and error in entrepreneurship has been reduced to a level that ordinary people can afford. The cost of a failure has shifted from “losing all savings” to “a weekend.”

This is the truly exciting part of the AI era—it’s not about a few geniuses creating more powerful things, but about enabling the majority of ordinary people to participate for the first time.

4. A few suggestions for entrepreneurs in the AI era

After these two days, if I were to give specific advice to entrepreneurs who are on their way or preparing to start, I would say these points.

First, stop asking “what can AI do,” and start asking “what has been bothering me lately.” Products that start from technology and then reverse-engineer applications will always lose to those that start from real pain points and then develop technical solutions. Your advantage isn’t AI—since everyone has access to AI—your advantage is your deeper understanding of specific pain in a particular scenario.

Second, cross-industry experience is an asset, not a disadvantage. If you are a lawyer, doctor, teacher, nurse, decorator, or used-car dealer, don’t think that “not understanding technology” is a problem. You have something that pure technical backgrounds lack: industry intuition. AI cannot learn this, and engineers cannot acquire it just by studying. Only your years in that industry can give you this insight.

Third, get your product out first, then scale. Don’t follow the old path of “working in secret for a year before launching.” You can have an idea today, build it tomorrow, and release it the day after to gather feedback. Let the market tell you whether it’s worth doing, rather than relying on your own emotional attachment.

Fourth, warmth is the ultimate moat. All the most touching products on-site share a common trait—they make people feel that “the creator genuinely wants to help others.” Brain-controlled wheelchair, the middle schoolers’ note tool, every project that catches the judges’ eyes—this is true for all. AI can help you write code, make PPTs, generate content, but it cannot replace empathy. When everyone is using the same tools, your understanding of people becomes your only differentiator.

Fifth, execution is everything. None of the winners in this competition were those with the most dazzling ideas; they were those who truly got their projects running within 48 hours. The biggest illusion of the AI era is that because barriers are lower, everyone thinks they can start anytime. But the truth is: most people will never actually start.

5. Final thoughts

As a lawyer who has long handled cases in new economy and Web3 fields, I came to Zhangjiang specifically to watch this event because my clients—entrepreneurs in AI and Web3—are the main players in this transformation.

What I saw over these days is a group of extremely smart, passionate people who are finally mastering the tools of creation. Many of them will produce things in the next few years that we can’t even imagine today.

But from the cases I’ve handled, I also see the flip side: the risks on the entrepreneurial journey rarely appear suddenly at a single point. They grow quietly through daily decisions—product feature design, marketing wording, business model adjustments, partner choices. Each small decision seems insignificant on its own, but over time, they can lead to an unforeseen situation.

The more cases I handle, the more I realize: lawyers shouldn’t just be the ones helping you clean up after a problem occurs. In this era of weekly iteration, what entrepreneurs truly need is someone who walks alongside them, warning them before they hit the bend—someone who helps them navigate the long road, not just waits at the finish line.

Special note: This article is an original work by Lawyer Shao Shiwei. It reflects only the author’s personal views and does not constitute legal advice or opinions on specific matters.