Equinox Provides Near Real-Time Fitness Insights Across 80 Clubs Using AWS

Founded in 1991, Equinox is a luxury lifestyle and fitness company that has over 100 clubs worldwide. The company approaches fitness holistically, focusing on movement, nutrition, and regeneration. Premium, integrated experiences are a pillar of the company’s offerings.

“Our technology has always been focused on providing the best experience to our members,” says Eswar Veluri, chief technology officer at Equinox. “We focus not just on the time that they spend at the club but even on their time outside of our clubs.” To deliver connected experiences, Equinox has a mobile application, which its members use to locate nearby clubs, sign up for classes, and track their fitness goals.

Delivering fitness insights to members who sign up for cycling classes is one of the company’s core connected experiences. To do this, Equinox gathers data from its cycling equipment using MQTT protocol. Before using AWS IoT services, the company stored its data in an edge server and processed it using custom code. This setup, however, impacted the member experience—delaying the delivery of fitness insights by 2–3 hours.

Equinox wanted to rearchitect its stack so it could deliver biometric data to members in near real time. “We wanted to come up with a solution that was light on the client and more cloud based so we can process the data as soon as we get in and in a more efficient way,” says Aneesh Pillai, engineer at Equinox. So, the company developed a proof of concept to see if it could gather data from its equipment using AWS IoT Core, which easily and securely connects devices to the cloud.

Stability and speed were two key factors Equinox considered. The company also wanted to free its developers from managing complex integrations and provisioning hardware. “We did not want to be held back with operational challenges,” says Sindura Nallapareddy, senior director of engineering at Equinox. “We wanted to focus on the possibilities that we could create, including fitness challenges in our studios and across our clubs.”

Equinox validated that AWS IoT Core can support device connections with MQTT protocol, helping its developers seamlessly onboard the existing devices installed on its cycling equipment. “Using AWS IoT Core, we don’t have to worry about how we connect with the backend,” says Pillai.

For data processing, Equinox uses Amazon Data Firehose to reliably load near real-time streams into data lakes, warehouses, and analytics. The company created two data streams—one for collecting biometrics and another for collecting equipment data, such as battery power status. By using Amazon Kinesis Firehose, Equinox can scale to process 50 MB of data from each class on a daily basis.  

By processing its data in the cloud, the company has improved the accuracy and breadth of fitness insights it delivers to its members and personal trainers. “We were able to create additional logic that can calibrate the number of calories burned, miles cycled, and energy expended. And it’s all available in near real time,” says Pillai. Further, Equinox can schedule maintenance and repairs for its workout equipment in a timely manner.

Protecting its members’ data is a priority. To securely collect user datagram protocol (UDP) packets from the cycling equipment’s edge devices, Equinox established a separate network for transmitting MQTT signals. Member data is then stored in the cloud in Amazon Simple Storage Service (Amazon S3), an object storage service built to retrieve any amount of data from anywhere.

Equinox also developed an application that its developers use to centrally manage software updates across its bikes with AWS IoT Greengrass, an open-source edge runtime and cloud service for building, deploying, and managing device software. “The beauty of AWS IoT Greengrass is that it’s plug and play,” says Pillai. “Integrating the service into our architecture only took 1 month.”

Equinox rearchitected its stack on AWS in less than 6 months. Since then, the company has improved the stability of its services and onboarded 80 of its clubs, nearly doubling the number of clubs it supported before. “We’ve built a platform that will last us,” says Veluri. “Now, we can continually build
new features that will excite our cycling members. We don’t need to worry if our data is good or not, and our engineers can stay engaged from a product standpoint.”

Now that Equinox has enhanced its data processing, the company is beginning its AI and ML journey. To accelerate time to innovation, Equinox’s developers are experimenting with Amazon SageMaker AI, which gives them the ability to build, train, and deploy ML models—including foundation models—for any use case with fully managed infrastructure, tools, and workflows.

Providing more personalized services and predictive equipment maintenance recommendations is one of Equinox’s goals. “The confidence that we have in our data has unlocked so much potential,” says Veluri. “We can explore club-to-club challenges, cycling milestones, and individual competitions. There is a lot of excitement across the company to uncover and unlock even more experiences.”