The first AI agent of Twin is an agent for receiving Qonto’s customer invoices

When Twin come out of insole In January 2024, AI agents were more a theoretical concept than reality. Today the Paris -based company launches an automation agent in partnership with ArrangementStarting Fintech, which offers business bank accounts to more than 500,000 customers across Europe.

If you want to automate repetitive tasks, there are already several ways to deal with these problems. Some companies use API based with no low code automation code or RecordS Others rely on RPA software, such as UipthS

With its team of nine people, Twin believes that there is a much more effective way of dealing with automation. As you might guess, it included artificial intelligence and models for using a computer.

The invoice operator, the first Twin product designed for Qonto, is a good example of why it makes sense to use artificial intelligence. Qonto processes millions of invoices per month. And customers spend a few hours a month collecting invoices and uploads them to Qonto.

In the last three months, Twin has created an automatic invoices that can accelerate this process. When users start an invoices, Twin first retrieves the list of transactions with missing invoices. It then shows the list of services that must be accessible to invoices to the browser window showing the agent’s actions.

If you need to sign in to download invoices, the browser stops and asks you to enter your credentials manually. After you do this, you can click on a button to let the agent continue your work.

The Twin invoices then automatically finds your list of past transactions, downloads invoices and attracts PDFs to transactions in your Qonto account.

“When you do this on Qonto’s scale, you generally have to cover a very, very long queue of services. Thousands, tens of thousands and soon hundreds of thousands of different services that all use,” said the co -founder and CEO of TWIN and Hugo Mercier CEO while demonstrating the product.

“And that would be completely impossible with RPA, because you will need to create a personalized script on a website and then every time the website changes, you will need to change the script,” he added.

As for API -based automation products such as Zapier, Mercier said it took 10 years on Zapier to support 8,000 applications on its platform. Twin already supports thousands of applications for its invoices only a few months after starting the product.

Behind the scenes, Twin runs a chrome -based web browser on a server. The launch uses the Cua (computer use of Openai. In fact, Twin was one of the 15 companies who had to try Cua in beta.

CUA is also the model that power the OpenAi operator, its Prosumer product that allows you to enter prompt to let an agent take action for you. In addition to better performance, Twin believes that it should be easier to use agents that look at the network for you.

“We have worked a lot to do the experience extremely simple. We are really heading to the end user, maybe people who are not comfortable for technology. They don’t have to promote or configure nothing. You just go into accounts, start it and it is moving to find the invoices,” Merseur said.

Following the extraction of invoices, Twin believes that there are many industries that could benefit from B2B agent applications. For example, agents can automatically manage orders for e -commerce company, classify the market catalog, or retrieve information about call center agents.

Twin sets a future in which AI agents become more expensive, faster and more accurate in a wide variety of tasks. Now let’s see if the start can turn the main platform of the agent that feeds the invoices operator into a product that developers can start using in their own applications.