Highlights:

  • Mendix 10’s major upgrade is the ML Kit, a toolkit that enables developers to integrate AI models into their applications.
  • Mendix introduced new features and updated its long-term AI roadmap, including developing Mendix Chat, a chatbot that will help developers implement software constructs more efficiently.

Mendix has recently given a sneak peek of a new version of its low-code development platform that will simplify creating artificial intelligence applications for businesses.

Mendix is introducing Mendix 10, which includes two AI chatbots to assist developers in their work. The company will reveal more information about the update at an upcoming event soon.

Mendix’s low-code platform enables businesses to develop applications with less effort than is typically necessary. Using a drag-and-drop interface, developers may assemble prepackaged software components into apps. When needed, the components can be extended with custom code.

After being acquired by Siemens AG for USD 730 million in 2018, Mendix became a subsidiary of the company. Since then, the software provider has reported expanding its client base, with over 4,000 enterprises using its services.

Mendix 10’s major upgrade is the ML Kit, a toolkit that enables developers to integrate AI models into their applications. The toolkit supports models based on ONNX, a commonly used open-source neural network format in the enterprise.

Mendix claims that the ML Kit provides access to numerous pre-trained AI models, and businesses can also use their custom models. The toolkit is compatible with various ONNX-compatible AI development frameworks, including PyTorch and Caffee2.

Applications and their incorporated AI models often operate on distinct servers, resulting in data transfer latency. Mendix claims that ML Kit allows businesses to run the neural network that drives an application within the application’s container, reducing latency.

The toolkit also promises to provide faster software integration. Integrating an application and its AI model on separate infrastructure often requires significant custom code, which can take weeks. However, Mendix claims its platform can accomplish this in just a few hours.

According to Amir Piltan, Mendix’s Senior Product Manager for AI, “Enterprises with sophisticated machine learning capabilities can easily incorporate their models into Mendix applications using the ML Kit. But for those earlier in the adoption curve, it is not necessary for enterprises to build models from scratch.”

Alongside the ML Kit, a pair of new AI chatbots is being introduced by Mendix. The company claims that these chatbots are intended to assist developers in creating applications more quickly.

The first chatbot can scan an application’s code base to verify if it adheres to software development best practices. This automated process allows developers to focus on other duties. According to Mendix, the chatbot can also assist users with less technical expertise in ensuring that the applications they create meet quality standards.

The other chatbot introduced recently assists developers in creating data validation workflows. These workflows are responsible for guaranteeing that a specific application component only accepts input that it is intended to process. For example, a data validation tool can prevent users from mistakenly entering text into a form only to accept numerical input.

Software errors may occur when a piece of code receives data it is not designed to handle. By employing data validation workflows, such errors can be avoided, resulting in increased reliability of applications.

In addition, data validation is a crucial component of software teams’ cybersecurity measures.

Various cyberattacks, including SQL injections, exploit the vulnerability of an application that accepts input it was not intended to process. Such malicious requests can result in software errors that hackers may exploit to access sensitive data. The dangers of such cyberattacks can be reduced by implementing data validation code to filter out malicious input.

Mendix presented its new features and updated its long-term AI roadmap. The company outlined its progress on developing a forthcoming chatbot, Mendix Chat, which will assist developers in implementing specific software constructs more efficiently. Additionally, the Siemens unit has plans to leverage AI for various other development tasks in the future.

Mendix Chief Product Officer Hans de Visser said, “Our next step will be the introduction of ‘Mendix Chat,’ a chatbot in the Mendix IDE that will guide developers on how to apply certain concepts or patterns. We are currently training a large language model based on sources drawn from Mendix Forum, Mendix documentation, and our support system.”