AI applications are taking root in the network edge where they can make a big difference for small businesses and large businesses alike.
On a recent episode of Zero Touch Telecom Live, Rakuten Symphony’s LinkedIn Live series, two of our experts went live from the KubeCon + CloudNativeCon in Salt Lake City, Utah to discuss how a stateful edge compute infrastructure with Kubernetes-powered orchestration and management can change the game for a quick serve restaurant (QSR) recovering from a surge in orders.
W. Brooke Frischemeier, Head of Product Management, and Padmarajan (Raj) Narayanan, Global Head of Presales & Solutions, with Rakuten Cloud shared a story that illustrated how a quick service restaurant can use AI to make operations smoother and customer experiences more personalized.
The story begins with a baseball game that goes into extra innings and when it finally ends, a flood of hungry fans head straight to the restaurant. Everyone is craving hamburgers, and within hours, the restaurant’s hamburger inventory is wiped out.
While inventory management systems assure store management that restocking was underway, the manager knows that to meet their daily sales targets, they need to pivot and sell more chicken sandwiches. This is where AI comes to the rescue.
Enter Johnny, a loyal customer who visits the restaurant twice a week and always orders a hamburger from the value menu. Thanks to the restaurant’s AI-powered loyalty program, they know Johnny’s preferences, ordering habits and his appreciation for a good deal.
Instead of waiting for Johnny to arrive and find his favorite menu item unavailable, the system gets proactive. Johnny receives a personalized notification on his phone: a coupon offering a discounted chicken sandwich meal if he stops by at 11 am before the lunch rush.
When Johnny walks into the restaurant the next morning, the loyalty system recognizes his phone and activates the smart signage. The digital menu boards prominently displays chicken sandwich deals, aligning perfectly with Johnny’s special offer. Even more impressively, the counter worker greeting him sees Johnny’s name and coupon details on their screen. “Hey Johnny! We’ve got that chicken sandwich deal ready for you—would you like to try it today?”
This seamless blend of technology and personal attention doesn’t just help the restaurant sell more chicken sandwiches when hamburger inventory is low—it strengthens Johnny’s connection to the brand. He feels valued, and the personal interaction turns what could have been a frustrating experience into a positive one.
AI didn’t just drive sales in this scenario; it deepened customer loyalty.
By using AI to anticipate inventory needs, personalize offers, and enhance customer service, this QSR turned a potential inventory crisis into a loyalty win. What’s needed to make this a reality, Frischemeier said, is an edge compute system powered by Kubernetes.
When decisions need to be made in real-time—like recommending menu items or adjusting inventory to boost sales—it’s impractical to rely on centralized cloud computing.
Sending data to the cloud and back introduces transport latency, which can disrupt critical operations. By processing data locally, restaurants can analyze customer habits, manage inventory and optimize smart signage immediately without waiting for cloud-based responses.
Of course, added Narayanan in his conversation with Frischemeier, this approach isn’t just for restaurants; it applies to any business operating remote locations disconnected from traditional data centers. Whether it’s retail, logistics or manufacturing, edge computing provides the onsite decision-making power necessary to deliver efficient and personalized experiences.
Deploying and managing edge computing infrastructure across thousands of locations might seem overwhelming. Each site could involve multiple applications handling point-of-sale (POS), inventory management, kitchen operations and security among other functions. Keeping these applications up-to-date, secure and functioning seamlessly across a sprawling network is a significant challenge. That’s where Kubernetes shines.
Kubernetes orchestrates the deployment and management of containerized applications, simplifying what might otherwise be a logistical nightmare. With Kubernetes, businesses can centrally manage thousands of edge locations, ensuring updates, patches and new features are rolled out consistently across all sites.
Policy-driven automation takes Kubernetes capabilities even further. Automating routine tasks—like updating POS systems or patching security applications—removes human error and increases operational efficiency. For example, when a new software patch is available, policy-driven automation ensures that it’s applied uniformly across all edge devices without manual intervention. This keeps systems secure and minimizes downtime, even across large enterprises with extensive remote operations.
Many edge applications are stateful, meaning they need to retain data about ongoing operations. Whether it’s a shopping cart remembering what items a customer added or a banking system tracking transactions, maintaining state is essential for these business-critical applications. Traditional Kubernetes lacks native support for stateful applications, but this limitation is overcome by integrating software-defined storage (SDS).
SDS provides the persistent state needed for cloud-native platforms. It ensures that edge applications can operate reliably, remembering customer preferences, inventory changes, and other critical data even during system updates or disruptions.
For the full version of the conversation with Narayanan and Frischemeier from KubeCon, visit the Rakuten Symphony Zero Touch Telecom LinkedIn page.