How Sportivity used agentic AI-supported development to build an IoT solution faster

Sportivity may not immediately ring a bell, but there is a good chance you use its app every week when checking in at your gym.

With more than 2 million daily users in the Netherlands, Sportivity is one of the country’s largest software providers for gyms. The platform supports processes such as membership administration and gives gym members access to different features through the Sportivity app.

Blue Green Solutions previously helped develop this Mendix application. When Sportivity wanted to create a new IoT connection with smart fitness equipment, it once again asked Blue Green Solutions for support. This time, however, a different technology proved to be a better fit.

This case shows how Mendix and agentic AI can complement each other within the same application landscape.

Connecting the Sportivity app to smart fitness equipment

Sportivity wanted to connect its existing app to fitness equipment designed for pelvic floor training. When a gym member scans the barcode on a training chair, the Sportivity app and the chair need to exchange the right data immediately.

This created a typical IoT challenge: how do you enable an existing application to communicate quickly and securely with physical devices located at different sites?

A central communication layer was still missing. This layer needed to support fast and reliable communication while also allowing Sportivity to manage the chairs remotely. The goal was to centrally update settings, training programmes and screens instead of configuring each device separately on location.

The challenge therefore had two parts: how should Sportivity connect the app and the chairs, and which technology was the best fit for this specific IoT use case?

High-code as a communication layer alongside Mendix

Because the Sportivity app was built with Mendix, the new server could also have been developed with low-code. However, Blue Green Solutions recommended using high-code because it was better suited to this specific function.

The server needs to receive, process and forward information quickly. High-code is a proven approach for this type of task and provides more freedom to design the data processing and communication in a focused way. The existing Mendix application did not need to be changed or replaced.

Blue Green Solutions developed an MQTT server to manage the communication between the app and the chairs. MQTT is a widely used communication protocol for IoT applications.

When a user scans the barcode, the app sends data to the server. The server processes this information and sends the required data back to the chair. The same infrastructure also makes it possible to centrally manage the connected devices and change their configuration remotely.

This creates an IoT architecture in which the Sportivity app handles the user interaction, while the MQTT server manages communication with the physical equipment.

Agentic AI-supported development

Blue Green Solutions used agentic AI to support the development of the MQTT server. AI helped with tasks such as writing code, connecting existing software libraries and building more complex components, including a two-factor authentication flow.

This allowed the team to show a working version sooner and collect feedback from Sportivity earlier in the development process. The estimated development time decreased from eight or nine weeks with Mendix to around six weeks with AI-supported development.

Developer expertise remained essential. Especially in an IoT solution, where different systems and physical devices continuously exchange data, architecture, security and reliability require careful attention. AI-generated code is not automatically secure, well structured or scalable.

Translating customer requirements into technology, making technical decisions and incorporating feedback also remain human responsibilities.

To ensure that the new server runs securely and reliably, the server and its data are hosted through Nodge. The platform performs security checks and provides secure European hosting for applications developed with AI.

Central management through the existing Sportivity app

Thanks to the new connection, Sportivity can:

• configure the chairs remotely;
• centrally manage training programmes;
• configure screens remotely;
• exchange data between the chairs and the app.

As a result, Sportivity no longer needs to configure every chair separately on location.

The existing Mendix application and the new MQTT server now form one application landscape. Mendix supports the processes and user-facing features within the Sportivity app, while the high-code server handles the fast exchange of data with the chairs.

This case shows that organisations do not need to use the same technology for every application or feature. That is why Blue Green Solutions looks beyond the existing technology stack. We also consider which technology best fits the use case, the existing systems and the organisation’s goals.

Which technology fits your use case?

Are you already using Mendix, but unsure whether low-code is also the best choice for your next application or feature? We can help you assess your processes, technical requirements and future plans.

Together, we can choose the solution that best fits your organisation and existing systems.

Get in touch to explore which technology can help your organisation move forward.