The structure collects a seed of $ 4.1 million to convert uniceded web data into ready -made data sets for institutions

A start -up company based in Brooklyn aims to one of the most famous pain points in the world of artificial intelligence and data analyzes: the process of preparing painstaking data.

structure He came out of the ghost status today, announcing its general launch, along with $ 4.1 million in the financing of seeds led Bain Capital VenturesWith participation from 8VCand Integrated projects And strategic investors owners.

The company platform uses a special visual language model called Drain To automate, clean and structure data – a process that usually consumes up to 80 % of the time of data scientists, according to industry surveys.

“The size of the information available today has exploded,” said Ronak Gandy, co -founder of Structify, in an exclusive interview with Venturebeat. “We have reached a large turning point in the availability of data, which is a blessing and curse. While we have unprecedented access to information, it is not very possible that it is very difficult to convert to the correct coordination to make meaningful commercial decisions.”

The Structify approach reflects the increasing focus on the industry level on the solution to what the data experts call “the bottle cervix preparation”. Gartner’s research indicates this Prepare insufficient data One of the basic obstacles to implementing successful artificial intelligence remains, with four out of five works to the foundations of the data needed to fully benefit from artificial intelligence.

How to convert data that works with artificial intelligence intelligence intelligence is widely hidden

In essence, structuralists allow users to create dedicated data sets by identifying the data scheme, choosing resources, and publishing artificial intelligence agents to extract these data. The platform can deal with everything starting from SEC files and LinkedIn definition files with news articles and specialized industry documents.

What is a behavior, according to Gandhi, is its model in the Dora model, which moves on the Internet as a person does.

“It’s high quality. It moves and interacts with things as a person does,” Gandhi explained. “So we are talking about human quality – this is the first and most important thing for the principles behind Dora. He reads the Internet the way a person desires.”

This approach allows the structural support to support a free layer, which Gandhi believes will help to achieve democracy access to organized data.

Gandhi said: “The way you think about the data now, it is a truly valuable object,” Gandhi said. “This really precious thing that you spend a lot of time expressing work, wandering and wrestling, and when you have, you love,” Oh, if someone will delete it, then I will cry. “

The Structify vision is to “collect data” – which makes it easily re -created if lost.

From financing to construction: How companies publish data sets dedicated to resolving the challenges for the industry?

The company has already seen adoption across multiple sectors. Funding teams use them to extract information from the floors of the stadium, and the construction companies convert complex geo -documents into readable tables, and sales teams collect organizational plans in the actual time of their accounts.

STARTCTCThe Bain Capital Ventures partner has highlighted this diversity in the financing advertisement: “Every company has ever worked with it has a handful of data sources that have severe pain and great pain to work with it, whether it is buried numbers in PDFS, scattered across hundreds of web pages, hidden behind API Soap Enterprise, etc.”

The diversity of the customer base early in the global nature structure of the challenges of preparing data reflects. according to Techtarget ResearchData setting usually includes a series of intensive steps for employment: assembly, discovery, cleaning, disinfection, structure, transformation, and verification-all before any actual analysis begins.

Why does human experience remain decisive to the accuracy of artificial intelligence: the “internal” quadruple “verification system

The main distinction of structuring is the “Quartet verification” process, which combines artificial intelligence with human supervision. This approach addresses a decisive source of anxiety in developing artificial intelligence: ensuring accuracy.

“Whenever the user sees something suspicious, or we define some data as possible, we can send it to an expert in the case of the specified use,” Gandhi explained. “This expert can act in the same way (Dora), and move to the correct piece of information, extract it, and save it, then verify whether it is true.”

This process does not only correct data, but also creates training examples that improve the performance of the model over time, especially in specialized fields such as building or pharmaceutical research.

“These things are very messy,” Gandhi pointed out. “I never thought about my life, I will have a strong understanding of geology. But there, we, as I think, great power – to be able to learn from these experts and put it directly in Dora.”

When data extraction tools become more powerful, privacy concerns are definitely created. Strinchify has carried out guarantees to address these issues.

Gandhi said: “We do not authenticate, anything that requires login, anything that requires you to fail to feel the information – our agent does not do that because this is a source of concern for privacy.”

The company also gives transparency priorities by providing direct sources information. “If you are interested in learn more about a specific part of the information, you will go directly to this content and see it, instead of a type of old service provider where this black box is.”

The structure enters a competitive scene that includes both well -known players and other emerging companies that deal with different aspects of data preparation challenge. Companies like Almyxand Informaticaand MicrosoftAnd Plate All data preparation capabilities are provided, while many specialists have been obtained in recent years.

What distinguishes a structure, according to CEO Alex Reichnabach, is a mixture of speed and accuracy. The LinkedIn Publicity recently claimed by Reichnbach claimed that they had accelerated their “10X During the cost ~ 16X” agent by improving the form and infrastructure.

The company’s launch comes amid increasing interest in automating data operating in Amnesty International. According to Techtarget reportThe automation of the data preparation “is often martyred as one of the main investment fields for data and analyzes”, while increasing the increasingly enhanced data preparation capabilities.

How did you inspire frustrated data preparation experiments with two friends to revolutionize the industry

For GandHi, a structure makes problems that address the problems he faced directly in the previous roles.

Gandhi recalls: “The big thing in the founding story of the Temple is that they are a kind of personal and professional characters,” Gandhi recalls. “I was telling (Alex) the time when I was working as a data analyst and performing consulting and consulting operations, preparing these truly specialized data sets for customers – lists of all the fitness effects and their following standards, the lists of companies, what functions they publish, and museums on the eastern coast … I was spending a lot of time to coordinate them, rid them,“ all these jobs. ”

The inability to repeat quickly from the idea to the data set was particularly frustrated. “What made me do is that you cannot repeat and a kind of idea to the data set quickly,” Gandhi said.

The co -founder, Alex Reichnbach, faced similar challenges while working in an investment bank, as the data quality problems have hindered efforts to build models at the top of the organized data collections.

How to plan a structure to use seed financing of $ 4.1 million to transfer the institution’s data setting

Through new financing, form plans to develop its artistic team and establish itself as “the transition to the industries”. The company currently offers both free and paid levels, with institutions options for those who need advanced features such as local publishing or extracting very specialized data.

With more companies investing in artificial intelligence initiatives, the importance of high -quality organized data will increase. accident Massachusetts Institute Technology Review Report I found that four out of five companies are not ready to take advantage of obstetric artificial intelligence due to poor data foundations.

For GandHi and Structify, the solution to this basic challenge can open a great value across industries.

Gandhi said: “The fact that you can even imagine the world of creating data groups is to repeat a kind of mind in puzzle for many of our users,” Gandhi said. “At the end of the day, the stadium revolves around the ability to get this control and customization.”