36Kr teamed up with Zhejiang to create a series of special topics on "Her Power", trying to make a three-dimensional portrait of this group of new women in the workplace.
There are many female CEOs in the technology entrepreneurial circle, but there are only a handful of female CTO. The role of "controlling the company's technology and controlling the company's future" has discouraged many girls.
Zhiyi Technology CTO Wen is such a technical decision maker. Born in 1988, she has always been "someone else's child" in her parents' mouth: in 2005, when she was sent to eecs as a sophomore, she switched to artificial intelligence; Since then, Wen has obtained a master's degree in informatics from the University of Tokyo and a doctorate in artificial intelligence from Carnegie Mellon University in the United States.
Top female students return to China to start businesses and choose clothing industry. Before establishing Zhiyi Technology, Wen worked as a data scientist in Coursera (a large American open online course website).
But the two-point and one-line life made Wen realize that he was not suitable to be a "migrant worker".
It happened that her husband was exposed to the clothing industry while doing the project, which gave Wen the idea of returning to China to start a business. After market research, they found that fashion design, as an important driving engine of the industry, has some problems, such as low online data, large design data and poor structure, which leads to the low level of informatization of the whole industry.
Zhiyi Technology CTO Wen Miaomiao
Why not use artificial intelligence technology to empower the clothing industry?
In 20 18, Wen returned to China and founded Zhiyi technology with her husband Zheng Zeyu, one as CTO and the other as CEO. Zhiyi Technology has also become a veritable husband and wife startup.
At the beginning of the establishment of Zhiyi Technology, Wen also thought about the direction of technology and service. She said frankly: "At first, we wanted to do To C business. After half a year of trial and error, we decided to switch to the to To B SaaS business. "
Today, Zhiyi Technology has developed into a technology company serving the clothing industry with artificial intelligence technology. Committed to solving the problems of fashion trend mining, clothing style search and recommendation, fast fashion clothing design and order service, the products cover AI data, knowing money, easy selection, blast furnace and so on.
Wen is a strong supporter of AI data technology behind Zhiyi Technology.
"In clothing enterprises, designers rarely touch digital information, and many designers don't even know what styles their company sells. Zhiyi Technology helps designers accurately analyze their own brands and competing products through data analysis. For example, which elements are popular, and the proportion of price segments and categories. The data provides a more rational judgment standard, which can help designers grasp the changes of fashion trends and improve the explosion rate. " Wen said:
How to integrate technology into the needs of the industry For the Internet, technology can collect a large amount of data in a certain field, but how users can use data to complete efficient work is a difficult problem.
Just like the clothing industry, from the designer's point of view, pictures are a very effective carrier to convey information, and how to extract professional labels from images is the key. Therefore, the data processing method of seedling selection is label recognition based on image data. "Identifying image labels first requires a professional labeling system and needs to be combined with the industry."
According to Wen, Zhiyi Technology's label system is an intelligent system with more than 500 identification labels based on the existing mature label systems in the industry and through communication with a large number of designers.
Zhiyi technology AI data products
"For clothing data, we first need to have a large number of accurate training data in the clothing industry. In the long-term data accumulation process, Zhiyi Technology has more than 6.5438+million training data, and all training data are completely marked by a professional team of designers. At the same time, we have a very strict quality control system, which can make the accuracy of label recognition reach a relatively high level. "
With the ability of data collection, how to help designers in practical application? Wen Miaomiao explained: "Our AI algorithm can identify the design points on clothing and statistically analyze the changing rules of fashion trends. In this way, designers can predict the quality of styles through data when designing. "
Zhiyi Technology's collection, collation and analysis of clothing e-commerce data can help customers find the explosions on the market faster. At the same time, how to price, how to set colors, how to allocate SKUs, how to plan different colors and how to plan different categories can be seen at a glance on the platform.
With the rise of fast fashion and live broadcast, brands demand more and more styles and rely more and more on external design resources. Facing the huge fashion design market, Zhiyi Technology is constantly expanding the team size of the company, from simple software services to closed-loop service mode of design and supply chain.
"Up to now, the whale heart design team under Zhiyi Technology has signed nearly 100 professional in-service designers and cooperated with nearly 200 buyers to provide professional design support to customers. At the same time, in order to help customers provide more professional products and services, we will complete accurate matching through the user portrait of the merchant and the data portrait of the commodity, which can not only help designers face more customers, but also help customers find the right design more accurately. " Wen said: