SLA-as-Code and the Future of Network Reliability

28th Oct 2025
SLA-as-Code and the Future of Network Reliability—Q&A With Shon Lonkar Written by Paul Chaney Making next-generation wireless networks function for self-driving cars and emergency communication demands new methods of enforcing service-level agreements (SLAs). Traditional static policies and reactive monitoring are ineffective because these systems necessitate low-latency, dynamic optimization, and zero downtime. By embedding SLA-as-Code into network logic, organizations can programmatically control traffic flows and network behavior in real time. Shon Lonkar, a senior systems architecture engineer with more than 16 years of experience spanning 2G to 5G and now contributing to 6G development, leads efforts in network systems transformation. He played a pivotal role in advancing packet core network capabilities, particularly for SOS (911) and Wireless Priority Services (WPS) data and voice services. His innovations have resulted in multiple patents, including dynamic traffic management systems, orchestrator-based routing solutions using User Equipment Route Selection Policy (URSP), and predictive fault mitigation through closed-loop automation. In this exclusive Q&A, Lonkar discusses the core principles of SLA-as-Code, the technology stack that enables it, and how it is reshaping the future of network reliability for enterprises and critical services. Q: What is SLA-as-Code, and how does it differ from traditional SLA enforcement? Lonkar: SLA-as-Code converts static service agreements into real-time, programmable network policies. These embedded rules direct how traffic flows, whether for speed, security, or reliability, based on exactly what each application needs at any given moment. For example, the network automatically selects fast, high-bandwidth transmission paths when users access gaming applications. Meanwhile, the network directs banking traffic through protected and reliable network slices for security purposes. SLA-as-Code enables the network to make these adjustments automatically, without human intervention. The key difference between SLA and SLA-as-Code is in how each model manages service delivery. Traditional SLAs react to failures after they occur. In contrast, SLA-as-Code leverages artificial intelligence (AI) and automation to track network performance, user activity, and system status. It monitors the network in real time and detects and resolves potential issues before users notice, operating at speeds that human operators cannot match when handling large volumes of data. Q: Which technologies make SLA-as-Code possible, and what use cases benefit most from it? Lonkar: Implementing SLA-as-Code depends on several key technological components. For instance, managing radio access network (RAN) and core infrastructure depends on AI-driven orchestration systems, which serve as the base for this technology. The orchestrators utilize real-time telemetry data to monitor service-level indicators while making automatic adjustments to network policies, ensuring consistent results across all infrastructure layers. The 5G network benefits from orchestration through its ability to create dedicated network slices, enabling operators to allocate resources among different applications and customer groups. The network supports two distinct slices, serving various purposes: one for high-throughput consumer video traffic and another for ultra-reliable low-latency communications (URLLC) for autonomous systems. Using URSP in 5G technology allows devices to make application-specific routing decisions directly at the device level. Network slices operate independently to handle multiple applications running on the same device through separate paths that fulfill their specific performance requirements. The 5G core architecture introduces logical interfaces that improve control capabilities. Orchestrators can use these interfaces to connect with policy control and user plane functions, enabling real-time adjustments to routes, bandwidth, and latency based on current network conditions. The combination of these technologies, facilitated by a centralized orchestration framework, permits SLA-as-Code to function as an intelligent, adaptable, network-wide control system. Several critical use cases demonstrate the value of SLA-as-Code implementation. For mission-critical services like 911 and first responders, downtime is not an option. SLA-as-Code ensures that this traffic is prioritized instantly, with no delays, no manual intervention, and no compromise. Autonomous vehicle and machine-to-machine (M2M) communication systems require continuous, low-latency network access to provide consistent, real-time performance. Real-time optimization systems that analyze route coverage and network congestion patterns deliver benefits to these systems. Network-driven software update guidance for connected cars helps prevent service disruptions during peak usage times by identifying optimal update periods. Q: What are the main challenges in implementing SLA-as-Code? Lonkar: The greatest challenge is trusting the system to operate without human intervention, especially when lives or revenue are at stake. Operators need assurance that AI will not only act quickly but also correctly. That’s why adoption starts with low-risk areas, then scales as confidence builds. Vendor compatibility is another issue. Operators work with diverse platforms from multiple vendors, including Cisco, Ericsson, and Nokia, among others. These systems often have different architectures, making orchestration a complex process. Finally, the cost of implementing AI and automation at scale is significant. Scaling SLA-as-Code involves more than tools. It’s a strategic investment in long-term service agility and operational resilience. These challenges necessitate a phased approach, starting with specific network elements and then expanding to broader deployment. Q: How does SLA-as-Code improve network reliability and scalability? Lonkar: SLA-as-Code technology offers two primary advantages. First, automation reduces human error, the leading cause of configuration failures. The system helps teams manage changing network situations by adhering to established policies and procedures. Second, automated decision-making enables faster response times. Networks can reconfigure in real time to detect hardware breakdowns, natural disasters, or unexpected traffic surges. The system achieves true scalability through its agile operations, which do not require additional operational costs. Q: How does closed-loop automation contribute to the network of the future? Lonkar: The network of the future benefits from closed-loop automation because it supports continuous monitoring and performance optimization. The system continuously measures essential performance indicators to link them with network modifications, ultimately achieving peak operational performance. It detects performance degradation through its monitoring capabilities and performs automatic parameter adjustment, reverting to the previous setting if necessary. In my experience, orchestrator systems can automatically remove underperforming nodes from service while redistributing traffic until the issue is resolved. The system reintroduces the node back into service automatically after it recovers from its unhealthy state. This self-healing pattern serves as a fundamental requirement for 6G networks and future autonomous systems. Q: What trends and innovations are shaping the future? Which data and research support the SLA-as-Code approach? Lonkar: Tomorrow’s networks won’t just run themselves; they’ll get smarter with every connection. AI-driven orchestrators will combine live performance data with lessons learned across the organization to continuously refine how the network operates. Technologies such as URSP and application-aware routing also transform the user experience. Instead of optimizing by subscriber or device, networks can now optimize by application. That granularity is a significant leap forward. At the strategic level, there are efforts to centralize orchestration logic, allowing learnings and improvements from different teams to inform decisions across the network. It’s about creating a coherent, adaptable system capable of scaling intelligently. Research from the Institute of Electrical and Electronics Engineers (IEEE) and other leading organizations confirms the effectiveness of orchestration, AI, and network slicing in improving service performance. Companies like Dell Technologies are developing real-time policy enforcement platforms that align closely with the SLA-as-Code model. Studies published by the Association for Computing Machinery (ACM) show that AI-native orchestration significantly enhances dynamic SLA enforcement, while research from initiatives like 6G Flagship also point toward AI-native networks as the foundation of future mobile infrastructure. These findings validate the current direction and emphasize the urgency of investment and preparation. Building intelligent, trustworthy networks Automation isn’t just a necessity. It’s the cost of staying competitive. To remain resilient and future-ready, networks will require intelligent, closed-loop systems that learn from real-time feedback and adapt accordingly. These self-optimizing capabilities allow networks to operate with agility, minimizing downtime and optimizing performance. A centralized orchestration system further enhances performance data, innovation metrics, and risk indicators into a single platform. This integration empowers teams to respond quickly to shifting network demands while maintaining established performance parameters. As organizations advance toward automation, trust is earned through incremental wins. Phased deployments provide tangible proof of value while minimizing disruption. That approach builds internal confidence and accelerates adoption. Shon Lonkar About the Author: Paul Chaney is a seasoned writer, editor, and content strategist who helps businesses craft compelling, ethical marketing narratives through his consultancy, Prescriptive Writing. With a focus on clarity, authenticity, and responsible communication, Paul empowers organizations to tell their stories with purpose and precision. Connect with him on LinkedIn.