Liang Wenfeng 

Liang Wenfeng, born in 1985 in Zhanjiang, Guangdong Province, China, is a pioneering entrepreneur and innovator who has significantly influenced the fields of quantitative finance and artificial intelligence (AI). As the co-founder of the quantitative hedge fund High-Flyer and the founder and CEO of the AI company DeepSeek, Liang has emerged as a transformative figure in China's technological landscape.

Liang Wenfeng's early life and education

A deep-rooted emphasis on education marked Liang's formative years. Raised in a modest family, his father, a primary school teacher, instilled in him a passion for learning and problem-solving. From an early age, Liang demonstrated an aptitude for mathematics and engineering.

In 2007, Liang earned a Bachelor of Engineering degree in Electronic Information Engineering from Zhejiang University. He chose to remain at Zhejiang University to pursue a Master of Engineering degree in Information and Communication Engineering.

During this time, Liang worked under the mentorship of Professor Xiang Zhiyu and authored a thesis titled "Research on Target Tracking Algorithm Based on Low-Cost PTZ Camera," a work that laid the foundation for his subsequent interest in machine learning and automation.

Early career and entry into quantitative finance

While still a graduate student, Liang witnessed the global financial crisis of 2007-2008. Intrigued by the volatility of financial markets, he began collaborating with his classmates to gather data and explore how emerging technologies like machine learning could be applied to quantitative trading. This marked his initial foray into the intersection of finance and technology, a theme that would define much of his professional life.

After completing his graduate studies in 2010, Liang relocated to Chengdu, Sichuan Province, where he experimented with the application of AI across various industries. These years were filled with trials and setbacks, but they ultimately solidified his understanding of how AI could transform traditional sectors.

His first major breakthrough came in 2013 when he co-founded Hangzhou Yakebi Investment Management Co., Ltd. alongside Xu Jin, a fellow alumnus of Zhejiang University. This venture served as his formal entry into the financial industry and provided a platform for him to refine his ideas about AI-driven quantitative trading.

Two years later, in 2015, Liang and his collaborators established Hangzhou Huanfang Technology Co., Ltd., now known as Zhejiang Jiuzhang Asset Management Co., Ltd. The company focused on integrating advanced algorithms into investment strategies, paving the way for a new era of efficiency in financial decision-making.

In February 2016, Liang co-founded Ningbo High-Flyer Quantitative Investment Management Partnership with two engineering classmates. High-Flyer distinguished itself by its unique approach to portfolio management, utilizing mathematics and AI algorithms to make investment decisions.

High-Flyer achieved remarkable results by eliminating the need for traditional human portfolio managers, managing over 10 billion yuan in assets by 2019. Liang described the firm's mission as "enhancing the efficiency and transparency of China's secondary market," a goal that resonated with investors and regulators alike.

Founding DeepSeek and shift toward artificial intelligence

By 2019, Liang's focus began shifting from finance to artificial intelligence. He founded High-Flyer AI, a subsidiary dedicated to the research and development of AI algorithms and foundational applications.

Anticipating the growing importance of computational power in AI development, Liang began acquiring thousands of NVIDIA GPUs as early as 2021. At the time, many industry insiders dismissed this move as overly ambitious or even eccentric. However, it soon became clear that Liang's foresight was well-founded.

In May 2023, Liang announced an ambitious new initiative to develop artificial general intelligence (AGI), a milestone that many experts consider the "holy grail" of AI research. This vision culminated in the establishment of DeepSeek, an AI company funded primarily by High-Flyer.

Venture capital firms initially hesitated to invest in the project due to uncertainties about its short-term profitability, but Liang remained steadfast in his belief that AGI represented the future of technology.

DeepSeek adopted a unique hiring strategy, prioritizing talent and passion over conventional experience. The company often recruited recent graduates from top Chinese universities, valuing their innovative perspectives and willingness to tackle complex challenges. This approach allowed DeepSeek to assemble a diverse and dynamic team capable of pushing the boundaries of AI research.

Innovations and impact of DeepSeek

Under Liang's leadership, DeepSeek quickly established itself as a formidable player in the AI industry. In May 2024, the company released DeepSeek-V2, a cutting-edge AI model that delivered exceptional performance at a fraction of the cost of its competitors.

This release triggered an AI price war in China, forcing major tech companies like ByteDance, Tencent, Baidu, and Alibaba to reduce the prices of their own AI models to remain competitive. Despite its low pricing strategy, DeepSeek maintained profitability, setting it apart from rivals that incurred significant losses.

In December 2024, DeepSeek launched its most advanced model yet, DeepSeek-V3, which featured an astonishing 671 billion parameters. Remarkably, the model was trained in approximately 55 days at a cost of under $6 million, a feat that highlighted the company's efficiency and resourcefulness.

Benchmark tests revealed that DeepSeek-V3 outperformed models like Llama 3.1 and Qwen 2.5, rivaling the capabilities of GPT-4o and Claude 3.5 Sonnet. By January 2025, DeepSeek's AI Assistant, powered by the V3 model, had surpassed ChatGPT as the top free app on the iOS App Store in the United States, a milestone that underscored the company's global impact.

DeepSeek's success has been hailed as a game-changer in the AI industry, challenging the dominance of American models and highlighting China's rapid technological advancements. The company's efficient use of resources has also demonstrated the limitations of U.S. sanctions aimed at restricting China's access to advanced AI chips and other critical technologies.

Philosophy of open source and innovation

A philosophy of open-source collaboration guides Liang Wenfeng's approach to AI development. He believes that in the realm of disruptive technologies, closed-source methods serve only to delay progress temporarily. DeepSeek has committed to open-sourcing all its models, fostering a culture of shared innovation and collective problem-solving. This commitment has earned the company widespread acclaim and positioned it as a leader in the global AI community.

Liang has also emphasized the importance of China transitioning from being a follower to a leader in technological innovation. He argues that China's challenges in innovation stem not from a lack of capital but from a deficiency in confidence and effective talent organization. By focusing on foundational technologies rather than immediate commercial applications, DeepSeek aims to contribute meaningfully to global advancements in AI.

Engagement with government and industry recognition

Liang's contributions to technology and innovation have not gone unnoticed by the Chinese government. On January 20, 2025, he was invited to attend a symposium hosted by Premier Li Qiang in Beijing. As an industry expert, Liang provided valuable insights and suggestions on the 2024 government work report, reflecting his growing influence in shaping China's technological and economic policies.

Through his leadership at High-Flyer and DeepSeek, Liang has advanced China's position in the global AI landscape and also challenged conventional norms, advocating for open-source development and efficient resource utilization.