news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

Reported by Wu Xiaoyu, a reporter of 21st Century Business Herald and trainee reporter Jiao Wenjuan

He Xiaopeng, CEO of Xpeng Motors, announced in his opening letter for the Year of the Dragon in February this year that Xpeng plans to recruit 4,000 people and invest 3.5 billion yuan in AI technology research and development.

Recently, Xiaopeng Motors' intelligent business has welcomed another key talent. 21st Century Business Herald reporters learned from multiple independent sources that Suo Kunlei, who has a background in AutoNavi, has joined Xiaopeng Motors and is responsible for the product experience of Xiaopeng's smart cockpit. His position is senior director and he reports to Wei Bin, head of the Internet Center.

The new head of the Data Intelligence Center (DIC) is Liang Jiankun, who reports to He Xiaopeng and has been with Xpeng Motors for more than half a year. The center is Xpeng's technical middle office, responsible for IT system operation and maintenance, data, and information system management.

Suo Kunlei has been working in the geographic information service industry for many years. He joined AutoNavi Maps in September 2011 and served as a product manager. Maps provide key spatial information and navigation guidance for self-driving cars, ensuring that vehicles can reach their destinations safely and accurately. Therefore, talents with a map background are favored by car companies. AutoNavi Maps is also an important source of talent for Xiaopeng Motors. Chen Yonghai, vice president of product planning at Xiaopeng, and Wei Bin, head of the Internet Center, were previously vice president and product director of AutoNavi Maps respectively.

Liang Jiankun, the current head of Xiaopeng DIC, has nearly 20 years of experience in computer technology and the communications industry. He has worked in various positions in IT-related industries such as software development, hardware design, system integration, communications networks, and Internet operations. He calls himself a "veteran of Internet entrepreneurship."

Liang Jiankun is He Xiaopeng's computer science alumnus at South China University of Technology, and the two also worked together at AsiaInfo. AsiaInfo Technologies, known as the "China Internet Architect", was mainly engaged in software development and providing system integration solutions for telecom operators at that time. In 2002, after leaving AsiaInfo, Liang Jiankun worked for China Netcom Group and 21CN, serving as a technical expert and CTO of the Guangdong Branch. Since 2008, he has started a series of businesses in the Internet field, and has successively founded Changying Network and served as a technical partner of 3N, which mainly provides technical services to some Internet companies. After joining Xiaopeng, Liang Jiankun took over the job of Huang Ronghai, the former head of the Data Intelligence Center, who was transferred to the head of the human resources department in January this year.

In August last year, after Wu Xinzhou, the "soul figure" of Xiaopeng Intelligent Driving, left, Xiaopeng's intelligent business entered a dynamic adjustment period, with some old employees leaving and new employees joining.

According to incomplete statistics, nearly a year after Wu Xinzhou left, seven veterans of the Xiaopeng Intelligent Driving team left.

In November 2023 and January 2024, Parixit Aghera, head of the software department of Xiaopeng Motors, and Liu Langechuan, head of the Xpilot project, resigned and followed Wu Xinzhou to join Nvidia. Since January this year, Wang Tao, former head of visual perception of the US team, Liu Yilin, former senior director of autonomous driving products, Pengyue Zhang, head of the static group of the perception module, and Lin Yishu, head of the North American intelligent driving team, have left Xiaopeng one after another.

According to incomplete statistics from public reports, including Suo Kunlei and Liang Jiankun, Xiaopeng has 5 middle and senior management joining the company in terms of intelligence this year.

Recently, Zhang Yu, former head of perception at QINGZHOU Zhihang, joined Xpeng Motors. In March this year, Xianming Liu, a former senior machine learning engineer at Cruise, a self-driving company under GM, became the head of Xpeng's AI team, replacing Liu Langechuan and responsible for the development of XNGP. In June this year, Yuan Tingting, former head of Alibaba DAMO Academy's self-driving business, became the senior director of Xpeng Motors' self-driving products, replacing Liu Yilin.

Overall, at the critical moment of end-to-end development, Xiaopeng still needs fresh blood to fill the gaps in its intelligent driving team.

End-to-end technology continues to heat up this year, and the first car company to take the lead is Tesla. In May 2023, with the launch of FSD Beta v12, Tesla became the first car company to introduce end-to-end architecture in smart driving. After that, domestic car companies followed suit. In May of this year, Xiaopeng Motors launched a full update of the AI ​​Dimensity system, becoming the first car company in China to mass-produce large end-to-end models. In July, at the Xiaopeng Motors AI Intelligent Driving Technology Conference, the Xiaopeng XNGP Intelligent Driving XOS 5.2.0 public beta version was launched.

Xiaopeng's pioneering end-to-end large-scale model does not mean that other car companies have no chance to catch up. An intelligent driving expert who has been in the industry for more than 10 years told the 21st Century Business Herald reporter that he believes that when entering a new technology cycle led by Tesla, car companies cannot use traditional time to estimate the time it takes for new technologies to be produced. "Don't think that it takes us as long as someone else."

Weilai and Ideal hope to win the opportunity to catch up in the end-to-end technology direction by adjusting their organizations. At the end of last year, Ideal set up a separate team for the end-to-end model, which was placed under the "algorithm research and development" team and was led by Jia Peng. In June this year, Weilai's intelligent driving organizational structure was integrated from the original perception, regulation and control, and integration into a large model team and a delivery team.

Compared with Ideal and NIO, Xpeng, which was the first to implement end-to-end, made the latest organizational adjustments. In August this year, Xpeng adjusted its intelligent driving organizational structure, setting up three new departments: AI model development, AI application development, and AI performance development to split the original technology development department.

One view is that Xiaopeng has explored various solutions, including those with and without maps, and while it has accumulated a lot of experience, it also has a "heavy burden" to make adjustments.

However, an insider close to Xiaopeng Motors expressed another point of view to the 21st Century Business Herald reporter, that organizational adjustment is not the goal, but a means to rapidly develop end-to-end technology, and the ultimate goal is to compete to see which car company has the best end-to-end effect after installing it. He believes that other car companies adjusted their organizational structures before the end-to-end large model was implemented, "in order to speed up implementation." Xiaopeng chose to adjust after mass production was implemented, in order to pursue the end-to-end effect of installing it "further." After installing the end-to-end system, car companies compete on the system's ability to iterate quickly. If they cannot iterate every 3 to 4 months and the effect increases exponentially, it is "pseudo end-to-end."

Xiaopeng's goal is to become a global pioneer and popularizer of AI smart driving. Achieving end-to-end means that the time for transition from automatic assisted driving to fully autonomous driving will be greatly shortened. According to He Xiaopeng's plan, Xiaopeng Motors is expected to achieve "door-to-door" smart driving in the fourth quarter of 2024. In 2025, Xiaopeng will achieve smart driving in urban areas comparable to smart driving on highways.

However, the road from end-to-end to fully autonomous driving cannot be separated from the long-term investment of real money by car companies, including huge computing power and massive fleet data. In the first quarter of 2024, Xiaopeng's R&D investment was 1.35 billion yuan, and it plans to invest 3.5 billion yuan in AI-related technology research and development in 2024. In the first quarter of this year, Ideal Auto's R&D expenses reached 3 billion yuan, a year-on-year increase of 64.6%; Weilai invested 2.86 billion yuan in multi-dimensional research and development such as intelligent driving.

In terms of cash reserves, Xpeng Motors currently does not have an advantage. According to the financial report, as of March 31, 2024, Xpeng Motors' cash reserves were RMB 41.4 billion, while Ideal's were RMB 98.9 billion during the same period.

Selling cars is the core source of revenue for car companies. In terms of the annual delivery target of 280,000 vehicles, Xiaopeng's KPI in the first half of this year was only about one-fifth completed. In order to realize the ambition of AI intelligent driving, Xiaopeng's most urgent thing at the moment is to sell cars.