In technology, many entrepreneurs receive only one determining act. Liang Wenfeng, the founder of Deepseek, is already at his second.
The engineer, described by colleagues as an introspective, has for the first time left his mark in the Chinese investment world in the late years 2010, which was a hedge fund, which used models for artificial intelligence to achieve strong returns and attracted billions of dollars in capital.
Disturbed by profit and wary of the tightening of Beijing's tightening up on speculative trade, Mr Liang played in 2023. He gave money in artificial intelligence, gambles on AI chips and assembly a team to build China's answer On the Silicon Valley Front-Runner OpenAi.
Now, just two years later, Deepseek has increased the global technical landscape. This is what you need to know about Liang Wenseng.
He is a deep technical engineer. That places him in a row of other successful Chinese technical managers.
When Chinese technologists debated why the largest investors and technology companies in the country had not anticipated the rise of generative AI, many people pointed to a single perpetrator: the companies of China were obsessed with rapid returns in a fiercely competitive market.
Armed with those lessons – and supported by his own commercial writer – Mr. Liang made it clear that his ambitions are much beyond commercial applications.
His focus, he said, lies on what he sees when China's only real chance to catch up with the United States. That means that daring, idealistic fluctuations are taken with fundamental AI challenges. His primary ambition is to create artificial general intelligence or AGI – the elusive goal of building machines that can think and learn like people.
When Deepseek undermines his Chinese competition last year by offering his model at bargain prices, forcing larger rivals in their own price cuts, Mr. Liang rejected the meaning.
“To be honest, we didn't really care – it was just something we did along the way,” he said in a widely shared interview with 36KR, a Chinese technical outlet. “Offering cloud services is not our main goal. Our goal is still to reach Agi ”(Deepseek has usually stopped this week and did not respond to requests for comments.)
In his beliefs that superhuman artificial intelligence is around the corner, Mr. Liang a lot on the director of OpenAi, Sam Altman. But the similarities end there. Mr. Liang, a low-profile manager with a deep technical background in AI-engineering, fits more with the form of Pony Ma, a co-founder of the Chinese Tencent than the charismatic visionaries of Silicon Valley.
He started as a trader in Hedgefonds. And then he turned to pure ai research.
In many ways, Mr. Liang's career follows the most important shifts in the technological landscape of China.
His dissertation from 2010 at the University of Zhejiang started what would soon be one of the most popular topics in the Chinese AI: improving intelligent tracking algorithms for surveillance cameras.
Later, the hedge fund that he co-founded, buffered by regulatory pressure, which ultimately forced the closure of one of the most important investment products, according to Peter Alexander, director of Z-Ben Advisors, a market consultancy, which is investigating Chinese hedge funds.
“Between 2019 and 2023 they wanted this side project, so that their Ph.Ds felt like they had something to do and so Deepseek came out of it,” said Mr. Alexander.
“But it really went into overdrive when their primary investment product had to be closed in February 2024,” he said.
In a sense, it was China's action against the private sector that bumped Deepseek in the direction of long-term AI research.
He is willing to try things that other entrepreneurs will not do. He even hired illuminated Majors.
If you were to interview the AI experts from China who would deliver the first large generative AI breakthrough of the country, few would have chosen Mr. Liang. That included the Chinese government.
Deepseek was a private company without clear support from the state, not large alliances and none of the institutional weight of players such as Baidu, the search giant. In a system that preferred insiders, Mr. Liang was not one.
Yet there is a precedent for that. Some of the most disturbing technology companies in China – Huawei, Alibaba, Bytedance – started outside the spotlight, just to define their industry again.
Mr. Liang's approach was just as unconventional as the rise of his company. He has emphasized intellectual exploration of pure routine. His recruitment philosophy is equally unorthodox – the technical teams of Deepseek are accompanied by literature lovers to refine the AI models of the company.
“Everyone has their own unique journey and brings their own ideas, so it is not necessary to push them,” he said in the 36KR interview.
In a technical culture defined by debilitating hours and hierarchy, who bordered Bohemian's prospects. Nevertheless, Mr. Liang insists that change is necessary if China wants to lead in frontier AI innovation.
“When Chatgpt came out, the entire industry in China was missing the confidence to pursue limit innovation,” he said.
“Innovation starts with confidence – and we often see that more from young people.”
Alexandra Stevenson Made reporting of Hong Kong contributed.
