Latest news with #EngineeringLeader
Forbes
01-08-2025
- Business
- Forbes
The Five-Year Tech Strategy Is Dead: Here's What To Do Instead
Rupesh Dabbir is an Engineering Leader at Google with over a decade of experience building highly scalable systems in the cloud. Not long ago, a technology leader's proudest achievement was a meticulously crafted five-year strategic plan. It was a document of certainty—a detailed roadmap rendered in Gantt charts and milestones, promising a predictable, linear path to the future. Today, that same plan is an anchor, chaining an organization to a world that no longer exists by the time the ink is dry. The blistering pace of innovation in AI has rendered traditional long-term planning obsolete. When foundational models that redefine entire industries emerge in months—not years—a rigid roadmap isn't a guide. It's a liability. Leaders who cling to the illusion of long-term predictability will be outmaneuvered by those who treat strategy not as a map but as a compass. To thrive in this era of constant, radical change, we must abandon the five-year plan and adopt a new set of principles built for volatility. Set A Direction, Not A Destination The fatal flaw of a classic roadmap is that it presumes you know the best route. In today's environment, that's a dangerous assumption. A better approach is to define a clear, unwavering North Star—a mission-oriented goal that provides consistent direction. Instead of dictating, "We will build Feature X using Technology Y in Q3," a North Star objective might say, "We will become the most trusted source for instant, personalized financial advice for millennials." This reframing liberates your teams. You're no longer telling your sharpest minds what to build; you're pointing to the mountain they need to climb. It empowers them to experiment with the latest AI tools—whether that's a new large language model or a novel data-synthesis technique—to find the most effective way forward, even if it was unimaginable six months ago. The "why" remains constant; the "how" becomes flexible. Build With LEGOs, Not Marble Enterprise architecture used to be built for scale. Now, it must be built for adaptability. A monolithic system—even one that scales—is a constraint. When a breakthrough AI model appears, integrating it into a rigid architecture could take years—by which time the technologyhas already evolved. For too long, we've built systems like marble statues: beautiful, fixed and inflexible. We now need to build like we're using LEGO bricks—standardized, modular and ready to be reassembled as needed. A modern tech stack must be API-driven and component-based. AI should be a swappable module, not a permanent fixture. This flexibility is the technical foundation of business agility. Forge Teams That Thrive On Chaos Finally—and most critically—leaders must foster team cultures that don't just tolerate uncertainty but thrive on it. Without a fixed roadmap, top-down directives fall short. Instead, leadership must provide context, tools and trust. A nimble team is one empowered to experiment, fail, learn and pivot fast. This means creating an environment where intellectual courage is celebrated and where engineers aren't penalized for trying something new. It means valuing the insight from a failed experiment as much as the result of a successful launch. Leaders must protect teams from the organizational inertia that stifles innovation. The goal is to build small, autonomous teams that move in rapid cycles, continuously iterating toward the North Star. Let's be clear: the era of certainty is over. The comfort of the five-year plan is gone, and it's not coming back. But for leaders willing to trade the illusion of control for the power of agility, this new world offers unmatched opportunity. The choice is simple: cling to the maps of a world that no longer exists—or grab a compass and lead the charge into the unknown. The future belongs to the navigators. Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Forbes
10-07-2025
- Forbes
The Evolution Of SON With Cloud-Native Architecture
Apeksha Jain is a Global Expert in Architecting Distributed Applications, Engineering Leader in Cloud-Scale Systems. Wireless networks have long faced the challenges of manually configuring, optimizing and maintaining thousands of base stations. This process is labor-intensive, economically inefficient and prone to errors. The inability to adapt in real time also makes it difficult to respond to issues like network congestion, meaning that base station faults go unresolved until manually identified. To address these challenges, the self-organizing network (SON) was introduced into 3GPP specifications in Release 8 in 2009. SON is a tool that provides self-healing and allows the network to detect and recover from failures on its own. This update enhanced operational efficiency through its automated functionalities, which include: • Self-Configuration: SON automates the setup of key parameters, such as initial neighbors, physical cell identifier (PCI) and tracking area code (TAC) for newly deployed base stations, ensuring compatibility with neighboring stations to prevent network interference. • Self-Optimization: SON can continuously monitor network performance and automatically adjust base station parameters to optimize resources, improve service quality and reduce interference. • Self-Healing: With SON, the network can automatically detect and resolve base station faults before they impact the network or users, leading to faster recovery and reducing operational costs. However, over the years, traditional SON architectures have faced several challenges, some of which can be overcome by using cloud-native SON. Let's explore those challenges, the role that cloud-native SON can play in addressing them and how to successfully implement cloud-native SON. The Challenges With Traditional SON Traditional embedded SON applications—also called distributed SON (D-SON)—deploy directly on network controllers. They enhance network management, but they also pose functional limitations in today's dynamic telecom landscape. When operating in isolation on a network controller, SON only has visibility into the base stations connected to a network controller, lacking a global view of the network. This limits SON's ability to implement network-wide load balancing needed to improve the overall efficiency of the network. D-SON can only load balance the user traffic between the base stations connected to the same network controller, which can also result in conflicting configurations across the base stations. As explained in an article on this type of SON architecture embedded within vendor-specific network controllers relies on proprietary interfaces. In multi-vendor environments, this design can make it difficult for SON functions to interoperate across different vendor equipment, which often leads to misconfiguration of neighboring base stations and inconsistent network behavior. Additionally, maintaining this solution may be operationally inefficient due to slow feature rollouts and resource constraints. Together, these limitations can prevent D-SON from leveraging advanced AI technology for predictive network configuration and management. The Shift Toward Cloud-Native SON Cloud-native SON is an architecture built on cloud infrastructure, typically delivered as a SaaS solution. It leverages the cloud's compute power and storage to process large volumes of network data in real time, enabling it to maintain a unified view of thousands of base stations across the network. As Nokia points out, this architecture is intended to enable automation, agility and seamless integration with other cloud-native network functions. The SaaS model can also support rapid deployment, continuous updates and operational efficiency to make network management more flexible and adaptive. Unlike traditional SON solutions, cloud-native SONs are developed by independent providers and built using standard APIs. A single SON application can, therefore, manage all base stations across the network and address interoperability issues between different vendor systems. With a holistic network view, cloud-native SON can prevent misconfigurations between neighboring base stations and support network-wide capabilities such as load balancing, improving overall performance and user experience. The cloud infrastructure is intended to provide a foundation for the integration of AI into SON applications. As noted by TeckNexus, AI-enhanced SON capabilities include predictive fault detection, intelligent load balancing and proactive energy optimization. This shift moves SON toward AI-supported decision-making, as it can offer more flexibility and scalability for emerging network requirements such as 5G. Adopting Cloud-Native SON While transitioning to cloud-native SON can offer substantial benefits, wireless operators need to be aware of the key challenges and navigate strategically to ensure a smooth transformation. Many operators lack in-house expertise in cloud-native tools and frameworks, making it necessary to upskill existing teams or recruit new talent with relevant experience. Beyond technical skills, the adoption of cloud-native SON also requires a cultural shift toward fast-paced Agile- and DevOps-driven practices. This operational mindset is essential for unlocking the full potential of cloud-native systems. Another major consideration is the integration of existing network controllers with cloud-native SON. Operators using traditional SON embedded in their network equipment often use proprietary interfaces. Moving to a cloud-native SON with a standard API interface requires a middleware layer in their network controllers, which can be a complex and resource-intensive task that requires careful planning and phased execution. There may also be concerns about data sovereignty, cloud security and compliance when offloading SON operations to the cloud. To address these concerns, organizations can adopt region-specific data hosting, end-to-end encryption and strict access controls. Regulatory compliance can be improved through robust auditing, and hybrid or private cloud models can provide greater control over sensitive data. By addressing these considerations in a timely and strategic manner, the operators can ensure the successful adoption of cloud-native SON. Conclusion As wireless infrastructure expands, there is a continuous push to accelerate SON adoption, streamline deployment and reduce operational costs, which is why cloud-native networks are gaining traction. Adopting cloud-native SON involves addressing key regulatory, technical and process challenges, but it can play a crucial role in the evolution toward AI-native networks. Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
