If someone asks you about the AI tools you use every day, you might immediately think of ChatGPT or Claude AI. Developed by the US-based AI firms OpenAI and Anthropic, these two AI tools are widely recognized.
However, China is not far behind in AI development. It has launched DeepSeek AI, which is a powerful tool gaining attention in the AI community.
DeepSeek AI stands out because it is advanced and easy to use, making it a popular choice for many people working in artificial intelligence. But do you know who developed DeepSeek AI? Let’s explore more about the creators behind this innovative tool!
From Entrepreneur to AI Visionary: Who is Liang Wenfeng?
Liang Wenfeng is a well-known Chinese entrepreneur and AI pioneer. He is highly regarded for his groundbreaking work in both finance and artificial intelligence. He was born in Wuchuan, Guangdong, to a family of teachers.
Wenfeng was a strong student and earned both a Bachelor of Engineering in electronic information engineering and a Master of Engineering in information and communication engineering from Zhejiang University.
His master’s thesis, “Research on target tracking algorithm based on low-cost PTZ camera,” centered on enhancing AI surveillance systems.
| Name | Liang Wenfeng (梁文锋) |
| Born | 1985 (Date not available) |
| Place | Wuchuan, Guangdong, China |
| Age | 40 (As of 2025) |
| Education Qualification(s) | Bachelor of Engineering in Electronic Information Engineering (2007) Master of Engineering in Information and Communication Engineering (2010) |
| Founder of | Ningbo High-Flyer Quantitative Investment Management Partnership Co. Ltd. (2016)DeepSeek AI Company (2023) |
Career Highlights
Early Experiments with Quant Trading: Liang’s entrepreneurial journey started back in the 2008 financial crisis when he and his classmates experimented with quantitative trading using machine learning technologies.
He started Hangzhou Yakebi Investment Management Co. Ltd. in 2013 and co-founded Ningbo High-Flyer Quantitative Investment Management in 2016, pioneering the integration of AI into financial markets.
High-Flyer Hedge Fund: It was the first of its kind in China and managed more than 100 billion RMB in 2021. Liang transformed the field with a unique approach to AI-driven investment strategies.
DeepSeek & AI Entrepreneurial Success: In 2023, Liang established DeepSeek, an AI company dedicated to creating artificial general intelligence (AGI).
The company has recently been making headlines in the world for its cost-effective AI models that have beaten competitors such as ChatGPT and Google’s Gemini.
DeepSeek achieved technological breakthroughs by strategically acquiring Nvidia GPUs before U.S. export restrictions, enabling the development of cutting-edge AI systems at a fraction of the cost.
What is Liang Wenfeng’s net worth?
According to the Hurun Global Rich List, DeepSeek CEO Liang Wenfeng’s net worth was approximately $4.6 billion as of 2025. According to Forbes, the CEO owns about 84% of DeepSeek and at least 76% of High-Flyer.
As of 2025, High-Flyer oversees approximately $8 billion in assets, with Liang’s share estimated to be around $180 million.
DeepSeek vs. Other AI: What Makes It the Best?
DeepSeek has emerged as a competitive player in the AI landscape, challenging established giants like OpenAI, Google, and Meta. Several factors contribute to DeepSeek’s competitiveness:
- Cost-effectiveness: DeepSeek’s development costs are significantly lower than comparable Western products, giving it an advantage in the global AI market. This cost-efficiency is achieved through innovative architecture and training methods.
- Architectural innovation: DeepSeek utilizes a “mixture-of-experts” (MoE) architecture, which optimises the training process by reducing parameter quantities and chip requirements. This approach allows DeepSeek to activate only a subset of its 671 billion parameters for each task, enhancing efficiency and performance.
- Performance in specialized areas: DeepSeek has shown impressive capabilities in technical tasks, particularly excelling in mathematics with a 90% accuracy rate. In some benchmarks, DeepSeek-R1 has even outperformed OpenAI’s models in mathematical reasoning tasks.
- Open-source approach: Unlike some competitors, DeepSeek takes an open-source approach, making it freely available and modifiable by the community. This accessibility can drive innovation and customization.
- Efficient resource utilization: DeepSeek’s development philosophy emphasizes efficiency, having been built using innovative training methods and less powerful hardware. This demonstrates how clever engineering can sometimes overcome resource limitations.
- Competitive performance: In general, knowledge benchmarks like MMLU and DeepSeek-V3 match GPT-4o’s performance (88.5% vs. 88.7%). While it may lag in some areas, like multimodal capabilities, it compensates with superior text-based reasoning.
- Rapid innovation: DeepSeek’s emergence has injected new momentum into the AI field, driving rapid technological iteration and innovation. This “catfish effect” has created substantial pressure on existing AI competitors, potentially accelerating advancements across the industry.
Scott Wu Net Worth: Devin AI Software Engineer, CEO of Cognition Labs
