Startup CodiumAI launches TestGPT in beta. This AI system should help developers to test their code.

New start-up CodiumAI enters the scene with TestGPT. This is a new solution that, as the name suggests, is based on a GPT model for generative AI. More specifically, it concerns OpenAI’s new GPT-4 model. TestGPT is not a debugging tool but a real test assistant capable of generating interactive tests for code. Such tests not only check whether an application is running, but also whether it exhibits the desired behavior.

Developing relevant tests is a time-consuming process on which a lot depends. The stability of an app is partly the result of a targeted and qualitative testing process. TestGPT should create relevant tests based on written code. It is then up to the developer to accept the tests, after which the code is subjected to them.

Integration in development environments

TestGPT is intended to complement other AI tools such as GitHub Copilot. The new solution is currently available as a beta. Over time, the AI ​​tool will be implemented directly in development environments.

Generative AI models have caught on with the general public since the launch of ChatGPT. Projects like TestGPT show that the capabilities of such models are not limited to text generation.