Last updated on April 8th, 2024 at 06:49 am
TestGPT, a beta version of Codium AI’s generative AI-powered code integrity solution, was made available today. The TestGPT model provides autogenerated software test suite suggestions for developers to speed up coding and bug scans, starting with Python and JavaScript, to assist developers in testing their code.
Codium assists developers in automating the crucial test creation procedure. The organization said it got $11 million in seed subsidizing to foster this man-made intelligence model.
What is The Cost of Getting The Software Wrong?
A tool like this has a lot of potentials. Software errors will cost a staggering $2 trillion in the United States alone in 2020, prompting many businesses to question the software’s quality. The cost of fixing errors rises as they spread throughout the software development life cycle. However, software testing is a time-consuming and laborious process.
Itamar Friedman and Dedy Kredo, who were in charge of product and research and development teams at Alibaba Cloud, had firsthand experience with these difficulties. They built Codium AI in 2022 after their experiences in software development, machine learning, and product management convinced them of the potential of AI large language models (LLMs) for software test validation.
TestGPT Benefits the Testers
Friedman, co-founder, and CEO of Codium stated, “As a developer, testing your code is important.” It not only helps you find bugs, but it also gives you valuable insight into your code and lets you know that you’re coding for a reason.
He stated that writing non-trivial test cases is laborious and frustrating. Writing tests can sometimes be difficult, but the alternative of allowing a bug to enter production can be disastrous.
Codium’s first tool is an extension for the integrated development environment (IDE) that makes it possible to iteratively generate tests and then modify code based on those tests’ results. This interaction with the developer encourages the developer to write better code while also helping the tool acquire a deeper comprehension of the code and produce tests that are more precise and meaningful.
The company says that developers who use Codium AI will find bugs and gain valuable insight into their code, which will make their software better and more useful. All over the world, beta testing companies are consulted for this particular purpose.
More Code Integrity For Faster Development
Like ChatGPT, Copilot, and other generative dev apparatuses, the TestGPT framework takes advantage of generative man-made intelligence models. In any case, TestGPT is centered around confirming the rightness of code versus the ideal detail, as per Friedman. It is intended to speed up development by enabling high code integrity.
Friedman stated, “It does a series of pre-and post-processing steps to ensure high-quality outcomes. It embeds testing best practices in its prompting process.”
Codium is currently available as an extension for well-known IDEs like PyCharm and Visual Studio Code. Plans call for additional collaborations and features, as well as support for additional IDEs and programming languages. The company stated that since its closed-alpha was released in January 2023, thousands of users have already installed Codium.
Codium AI intends to continue ensuring high code integrity by expanding and integrating into other phases of the software development life cycle. It is anticipated that this expansion will enable next-generation, test-driven development, test data management, auto-fixing of bugs, code improvement suggestions, and CI/CD integration.
