Who is Liang Wenfeng? To many in the West, he’s a complete unknown.
Liang isn't driven by the typical Silicon Valley playbook. His focus is on open source, universal access, and a deep-seated belief in the power of shared innovation.

My first encounter with Liang Wenfeng was in June 2018, at the Hangzhou headquarters of his firm, High-Flyer (幻方) Quantitative. I heard whispers that despite buying a new house, Liang was so consumed with developing trading strategies that he hadn't bothered to decorate. Instead, a tent stood pitched in his room, serving as his bedroom. Colleagues told me his life outside of coding was virtually non-existent.

As our conversation drew to a close, we discussed recruitment and performance reviews. Liang mentioned they'd hired PhDs from overseas, boasting backgrounds in computer science and mathematics. We were taken aback by the potential cost, especially when he explained they were seeking individuals capable of "cutting-edge and in-depth research in specialized fields."

"How do you evaluate them?" I asked.

Liang responded, "We look at their specific research areas, how they write papers, and their passion for research."

"Fortunately," I ventured, "with your firm’s scale and performance, commission-based income must be substantial.”

Liang paused, then clarified, “Well, actually, we only get commissions when clients redeem their funds. Right now, redemptions haven't been significant..."

The industry standard for performance fees in private equity typically follows the high-water mark method, accrued monthly. Liang's approach – only charging when clients withdraw – is remarkably client-friendly, but places immense pressure on the fund manager.

"So, how do you assess employee performance?" I pressed.

"We don’t really have performance metrics,” Liang stated simply.

"What if someone isn't producing results or contributing?"

Liang tilted his head, considering this. "When we hire, everyone participates in the selection process… My feeling is, if someone isn't contributing, it's because we haven't placed them in the right role."

“Do you have a grand vision?” I inquired.

“Hmm, perhaps to create a company that doesn’t charge performance fees or management fees” he mused.

My second meeting with Liang was on August 30, 2019, at the Private Equity Golden Bull Awards ceremony hosted by the China Securities Journal.

He shies away from public recognition under his own name. For instance, his substantial donations were made anonymously, signed "A Humble Little Pig."

That day, after Liang's speech, we had another informal chat. I remember circling back to the question of forgoing performance and management fees. Liang explained, "The idea is to build an open-source strategy platform, accessible to everyday investors."

Fast forward to December 16, 2024, a seemingly ordinary day just before the launch of DeepSeek R1, his groundbreaking AI model. Liang, who posts on social media maybe once a year, shared an article in his WeChat Moments. It was the preface he penned for the Chinese edition of “The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution,” a biography of Renaissance Technologies founder and "Quant King," James Simons.

The preface concluded with a powerful statement: "Whenever I encounter difficulties at work, I remind myself of Simons' words: ‘There must be a way to model price.’”

Simons was 44 when he founded Renaissance Technologies to pursue quantitative investing. Liang Wenfeng, now 40, has already dedicated 16 years to the field, achieving financial independence. Today, he's fully immersed in his technological passion, able to calmly – and detached from commercial pressures – drive forward fundamental research and exploration that requires vast resources. This has always been Liang’s focus. Open source, universal benefit, and efficiency gains – these principles permeate his journey through the peaks of quantitative investing and artificial intelligence. For years, he has maintained a low profile, almost to the point of obscurity… Yet, in his quietude, brilliance shines.

Liang Wenfeng: 30 Golden Quotes

Extracted from Public Reports, Including a Speech, a Preface, and Two Interviews with "Waves"

On Quantitative Investing:

1. Like many new technologies, quantitative investing was initially mocked. No one believed computers could invest like humans.
2. But Simons keenly foresaw that with advancements in computer technology, one day, the "impossible" would become reality. He made numerous early attempts, most unsuccessful, but he never gave up, believing time was on his side.
3. Whenever I encounter difficulties at work, I recall Simons' words: "There must be a way to model price."
4. In the information age, financial markets are fair and transparent. Human fund managers and computer models stand on equal footing, further clearing obstacles for the widespread success of quantitative investing.
5. Some ask, with quantitative investing, will humans still be needed in the future? Of course, we need a large number of programmers and researchers.
6. The progress of the quantitative private equity industry as a whole roughly follows Moore's Law, with investment capabilities doubling every 18 months. It's expected that in the coming years, the efficiency of China's stock market will further improve. This is an unstoppable historical trend.
7. When the market is efficient, you can simply buy index funds. Index funds represent true value investing, and the main body of wealth remains in the hands of ordinary people.
8. As hedge funds, our mission is to enhance the efficiency of China's secondary market.

On Artificial Intelligence:

9. High-Flyer, in a way, strengthened our confidence in technology-driven innovation, but it hasn't been without its challenges. We went through a long accumulation process. The outside world only sees High-Flyer after 2015, but we’ve actually been at it for 16 years.
10. We didn't intentionally become a "catfish" (disruptor), we just inadvertently became one.
11. (On stockpiling GPUs) Many might assume there's an unknown business logic behind it, but actually, it’s primarily driven by curiosity – curiosity about the boundaries of AI capabilities.
12. (Why DeepSeek currently focuses only on research and exploration?) Because we believe the most important thing right now is to participate in the global wave of innovation.
13. With economic development, China must gradually become a contributor, not just a free-rider.
14. What we lack in innovation isn't capital, but confidence and knowing how to organize high-density talent to achieve effective innovation.
15. Innovation isn't entirely commercially driven. It also requires curiosity and the desire to create.
16. In the face of disruptive technologies, moats built on closed-source approaches are short-lived.
17. Therefore, we invest in our team. Our colleagues grow in this process, accumulate a lot of know-how, and form an innovative organization and culture. That is our real moat.
18. Open sourcing and publishing papers – we don't actually lose anything. For technologists, being followed is a great sense of accomplishment. In fact, open source is more of a cultural behavior than a commercial one.
19. Giving is actually an extra form of honor. A company that does this also has cultural appeal.
20. We often say Chinese AI is one or two years behind the US, but the real gap is the difference between originality and imitation. If this doesn't change, China will always be a follower, so some exploration is unavoidable.
21. NVIDIA's leadership isn't just the effort of one company, but the result of the collective efforts of the entire Western technology community and industry. They can see the next generation of technology trends and have a roadmap in hand. The development of Chinese AI also needs such an ecosystem.
22. In the long run, we hope to form an ecosystem where the industry directly uses our technology and outputs, and we only focus on foundational models and cutting-edge innovation. Then other companies can build B2B and B2C businesses on top of DeepSeek.
23. What I often think about is whether something can improve the operational efficiency of society, and whether you can find a position you excel at within its industrial division of labor.
24. As long as the ultimate goal is to improve social efficiency, it's valid. Many things in between are just stages. Overly focusing on them will inevitably be dazzling and confusing.
25. DeepSeek (like the High-Flyer team) is also entirely bottom-up. And we generally don't pre-assign tasks, but rather have natural division of labor. Everyone has their own unique growth experience and comes with their own ideas, no need to push them.
26. Our hiring standard has always been passion and curiosity. So many people have some unusual experiences, which is very interesting. For many, the desire to do research far outweighs their concern for money.
27. The biggest draw for top talent is definitely solving the world's most difficult problems. And we are doing the most difficult things.
28. Using the business logic of the internet to discuss the future profitability model of AI is like discussing General Electric and Coca-Cola when Pony Ma (Ma Huateng, founder of Tencent) was starting his business. It's likely to be like "marking the boat to seek the sword" (using outdated methods).
29. China's industrial restructuring will be more reliant on the innovation of core technologies. When many people realize that past quick money may have come from luck of the times, they will be more willing to bend down and engage in real innovation.
30. In the future, there will be more and more hardcore innovations. When this society makes those who engage in hardcore innovation successful and famous, the collective mindset will change. We just still need a pile of facts and a process.