Zencoder just launched Amnesty International, which can replace quality assurance days within two hours

ZenCoderThe artificial intelligence coding, which was established by the series, series, Andrev ZitingAn agent of Amnesty International designed to automate comprehensive software test. This critical but slow step can often delay product versions for days or weeks.

The new tool represents the latest attempt by Zencoder to distinguish itself in the increasingly crowded artificial intelligence coding market, as companies are racing to automate not only to generate the code but the functioning of developing the entire software. Unlike the current artificial intelligence coding tools that focus mainly on writing code, Zentester targets the verification phase – guaranteeing the work of the program as intended before it reaches customers.

Filev said in an exclusive interview with Venturebeat: The CEO, who previously established the Project Management Company Your opinion And sell it to Citrix for $ 2.25 billion In 2021, he added: “Zentester does not only generate tests-it gives the developers confidence to the shipment by verifying that their symbol created from artificial intelligence or written on a person does what is supposed to do.”

This advertisement comes at a time when the artificial intelligence coding market is subject to quick unification. Last month, Zencoder captured the machinesAnged intelligence coding assistant with more than 100,000 downloads. At the same time, Openai has reached an agreement to acquire the Windsurf coding tool for About 3 billion dollars (The deal was completed in May). The moves emphasize how companies rush to build comprehensive Amnesty International development platforms instead of points.

Why the software test became the largest barrier in the development in which artificial intelligence works

Ziting The continuous challenge in software development is dealt with: long feedback rings between developers and quality assurance teams. In the home institutions environments, developers write a symbol and send it to the quality guarantee teams for the test, and they often wait several days to obtain notes. By that time, the developers moved to other projects, creating an expensive context switching when the problems are discovered.

“In a typical engineering process, after a developer creates a feature and sends it to quality guarantee, they receive notes after several days,” said Filev Venturebeat. “By that time, they already moved to something else. This context is replaced and backward-especially during the release-simple repairs can extend to a week’s ordeal.”

Early customer Club solutions group Dramatic improvements have been reported, as CEO Mike Servino said, “What took our quality guarantee team a few days taking two hours.”

The timing is particularly closely related as artificial intelligence coding tools generate large amounts of code. While tools like Copilot and Indicator They accelerated the generation of the code, and they also created new challenges to ensure quality. Filev estimates that if artificial intelligence tools increase the generation of the code by 10x, the test requirements will increase by 10x – overwhelming traditional guarantee operations.

How AI’s AI factors click on buttons and fill models such as human tests

Unlike traditional test frameworks that require developers to write complex texts, Ziting It works on ordinary English instructions. The artificial intelligence agent can interact with applications such as the human user – the buttons that link it, fill out forms, and move through the course of software work – while checking the health of both the fronts of the front end user and the back of the back.

The system is integrated with the current test frameworks, including theatrical writer and selenium, rather than completely replacing it. Filev said: “We never like people who give up things that are part of our DNA,” Filev said. “We feel that artificial intelligence should take advantage of the processes and tools already in the industry.”

Zentester provides five basic potentials: the developer -led quality test, the quality of quality guarantee to create a comprehensive test suite, improve the quality of the symbol created from artificial intelligence, automated test maintenance, and self -verification of continuous integration tubes.

The tool is the latest addition to the broader Zencoder platform, which includes coding factors for creating software testing factors and unit for basic verification. The company “Ribo re –Technology decomposes the entire code warehouses to provide context, while the error correction pipeline aims to reduce errors created from artificial intelligence.

The launch tightens competition in the artificial intelligence development market, where two players such as Microsoft’s are created Copilot Like the latest expatriates like Indicator They compete for the developer of the mind. Zencoder’s approach to building specialized agents for different stages of development with competitors who focus mainly on the generation of the code.

“At this point, there are three strong coordination products in the market: the degree of production: it’s the United States, the index, and the Wirkurf,” Filev said in an interview with it recently. “For smaller companies, it has become difficult and difficult to compete.”

The company claims superior performance in industry standards, as it reported 63 % success rates The bench has been checked Tests and about 30 % on the latest Swe’s multimedia standard Filev results say the best previous double performance.

Industry analysts note that the automation of the test to one side represents the next logical step of artificial intelligence coding tools, but successful implementation requires an advanced understanding of the logic of the application and the functioning of the user’s work.

What the buyer’s buyer needs to know before adopting artificial intelligence test platforms

ZenCoder’s approach provides both the opportunities and challenges faced by customers of institutions that evaluate artificial intelligence test tools. Company SOC 2 Type IIand ISO 27001 and ISO 42001 Certificates deal with security and compliance related to large organizations.

However, Filev admits that it justifies the institution’s alert. “For institutions, we do not call for the life development courses completely change,” he said. “What we call is AI-UAGMENTED, where they can now get a quick AI code review and an acceptance test that reduces the amount of work that must be done by the next party in the pipeline.”

Company integration strategy – Work within current development environments such as Optical studio code and Jetbrains Ides Instead of requesting the key system keys – it may appeal to institutions with applied tools tools.

The race to automate software development from the idea to publication

Zentester runs from Zencoder to compete for a greater share of software development work, as artificial intelligence tools expand beyond generating simple code. The company’s vision extends to full automation from the requirements to spreading production, although Filev is acknowledging the current restrictions.

“The next jump will be production requirements – everything,” said Filev. “Can you now activate it so that you can get the requirements of a natural language and then artificial intelligence can help you divide it, build architecture, build software instructions, build a review, verify that, and ship it to production?”

Zencoder Zentester provides three pricing levels: a free basic version, $ 19 per month user, and $ 39 per month user an institution option with distinguished support and compliance features.

As for the industry, it still discusses whether artificial intelligence will replace the programmers or simply make them more productive, Zentester suggests a third possibility: artificial intelligence that deals with arduous verification work while developers focus on innovation. The question is no longer whether the machines can write code – it is whether it can be trusted to test them.

Daily visions about business use cases with VB daily

If you want to persuade your boss at work, you have covered VB Daily. We give you the internal journalistic precedence over what companies do with obstetric artificial intelligence, from organizational transformations to practical publishing operations, so that you can share visions of the maximum return on investment.

Read with us privacy policy

Thanks for subscribing. Check more VB news bulletins here.

An error occurred.