CityFibre revolutionizes telecom with self-service stream processing

Picture this: You have a life-changing job interview over Zoom in an hour, but your internet connection is down. You spend time on hold with your broadband’s customer service to be told, “An engineer will get back to you.” When? Who knows.

But what if the platform your broadband service relies on could ensure that this particular issue never even happened? What if problems were detected and resolved before you ever knew they existed?

CityFibre, the UK’s largest independent full fibre infrastructure platform, is rewriting this story entirely.

CityFibre’s cutting edge infrastructure processes 100 million network events daily through their advanced Amazon MSK deployment. But raw technological power isn’t enough. The magic is in how they’ve democratized access to this data for their development teams to do something with it.

They found the answer in sophisticated event-driven architecture combined with AI-powered analytics; technology that has the potential to transform customer service proactively. But breakthrough results like this require the right engineering culture – one that gives developers the freedom to innovate with data.

Despite having comprehensive real-time network data, CityFibre faced a challenge: their development teams needed a highly efficient way to get hold of and transform the volumes of streaming data flows needed to build proactive customer applications.

The main challenge wasn't data availability; it was democratizing the data for developers. Development teams needed the ability to explore, filter, and transform streaming data without deep technology expertise (Apache Kafka), while maintaining governance across the organization.

CityFibre pioneers by putting data in the hands of those who need it, and making the latest technologies more widely accessible.

The development team uses the Lenses.iodeveloper experience to offer self-service capabilities for streaming data, transforming how developers can build the next generation of real-time applications:

• Instant data exploration: Teams explore streaming data using familiar SQL syntax, eliminating the need for specialized Kafka knowledge while accessing comprehensive real-time network insights.
• Rapid feature delivery with SQL Processors: Lenses SQL Processors allow teams to deploy applications that transform massive streams of data in 1 day rather than the traditional 7-day development cycle for custom coding.

"That’s been revolutionary." explains Adam.

Development teams have deployed 108 SQL Processors that process CityFibre's 8 million messages per day, joining topics and enriching data streams to be consumed by downstream applications. Critical network events are automatically identified and pushed to customer service systems in real-time.

With each processor saving 6 days of development time, this represents over 600 days of cumulative time savings annually. As CityFibre continues its mission to serve at least a third of the UK market and adds new streaming applications, so this time-saving compound effect will deepen its competitive advantage.

• Autonomous pipeline deployment: Teams independently deploy data pipelines that process Kafka data and push it to Snowflake via Kafka Connect.

"Lenses gives our teams the autonomy to build and deploy data pipelines, without requiring support by central IT teams," notes Adam.

This approach to simplify a sophisticated data streaming technology allows for rapid application development, focused on customer value. This also makes for happier teams; no one wants to be bogged down in complex engineering paradigms or the time pressures of a technology learning curve that just gets steeper.

The applications built on CityFibre's advanced platform are what make them stand out in the UK telecommunication industry, delivering remarkable customer service for their partners and end customers:

• 3-5 hours service restoration time: Their competitors take 3-5 days.

"With how quickly our engineering teams build streaming apps, our systems proactively dispatch engineers and fix problems before customers even know there's an issue," Adam says.

• Proactive customer communication: Applications automatically detect service issues and notify affected customers with specific problem details and exact repair times. That nightmare interview scenario? CityFibre's customers receive a notification that the issue has been detected and is being resolved in good time.
• Intelligent automation: "Our business applications run automatically, raise incidents, and dispatch engineering teams if required," says Adam. The system correlates issues across the network to identify broader problems, enabling intelligent resource allocation and faster resolution.

Instead of reactive customer complaints flooding call centers, CityFibre's applications proactively manage network health, dispatch resources, and transparently update the estimated time of resolution to customers.

CityFibre's leadership in technological advances extends beyond their streaming infrastructure.

They're pioneering AI-powered quality assurance through computer vision technology that analyzes engineer photos inside exchanges to confirm correct card connections and installations.

Working withDeepomatic, CityFibre has achieved zero quality-related field revisits, saving 55 minutes per operation verification while providing 100% quality assurance coverage compared to the previous 20% manual audit rates. This AI-powered visual automation prevents 3-4 incident visits monthly across their fiber network operations. In CityFibre's exchanges, a single installation error can cascade to affect thousands of customer services, making each prevented incident visit save potentially hundreds of hours of downtime and thousands of affected customers.

The company also leverages standard AI development tools like GitHub Copilot to automate mundane code completion tasks, while their enterprise data platform processes 2TB daily through AWS and Snowflake infrastructure. This foundation enables AI use cases to be deployed at scale much sooner than anticipated, supporting predictive analytics across CityFibre's ~4.4 million ready-for-service premises.

CityFibre is also driving industry standardization through TM Forum, creating standard APIs across telecommunications and leading an event-driven ecosystem initiative. This commitment to open standards enables faster application development across the telecommunications sector.

Looking ahead, CityFibre continues building AI-powered predictive applications that identify potential failures before they occur.

The next frontier: AI-led field operations enabling autonomous network management, representing the evolution toward generative self-healing networks.

CityFibre continues to transform telecommunications service through a combination of advanced event-driven architectures and AI-driven applications. By processing 8 million messages daily through self-service streaming data platforms and leading with an open approach to the latest technologies, they've enabled developers to build applications that proactively resolve customer issues in hours rather than days.

Their 108 SQL processors, each saving 6 days of development time per feature, demonstrate how simplifying complex technologies accelerates innovation. This work is helping CityFibre to set new standards for connectivity across the UK, operating the most efficient fiber platform while always striving to deliver the best customer service in the telecommunications industry.

As a result, customers no longer have to experience the stress and loss of precious time that traditional telecom service failures bring. Instead, they receive proactive communication, rapid resolution, and the premium internet service that modern life demands.