Amir Khan, President, CEO & Founder of Alkira – Interview Series

Amir Khan is President, CEO and Founder of Alkira. A seasoned networking visionary, he previously co-founded Viptela, a leader in SD-WAN acquired by Cisco, and has held leadership roles at major networking firms including Cisco, Juniper, and Nortel. At Alkira, he is pioneering new approaches to cloud-era networking and driving innovation at the intersection of AI and network infrastructure.

Alkira offers a Network Infrastructure-as-a-Service (NIaaS) platform designed to connect environments, sites, and users through a cloud-based enterprise network. The platform enables centralized management using familiar network controls, policies, and security frameworks. Delivered as a service, it is designed to scale on demand without requiring new hardware, software installations, or specialized architectural knowledge. Alkira’s solution is used by large enterprises, system integrators, and managed service providers.

What inspired you and Atif to found Alkira after successfully selling Viptela to Cisco, and how did you foresee the need for a unified cloud networking solution in the market?

After Viptela, we knew we’d build again, but the question was what. We looked at the networking industry and saw it hadn’t changed in years: it was complex, isolated, and heavily reliant on hardware. Our experience with Viptela taught us that there had to be a better way. So, we asked ourselves a simple question: why should building a network feel like a complicated puzzle with so many moving parts? Quite simply, it shouldn’t. So, we decided to break the old rules and build something completely different, something designed to not just improve networking but fundamentally transform it.

Now, that vision is a reality. We deliver Network Infrastructure-as-a-Service (NIaaS) built entirely in the cloud, offering the agility and scalability businesses need to focus on their core operations. Companies worldwide are using our intuitive NIaaS for on-demand scaling, without upfront hardware or complex integrations. It’s flexible and consumption-based, just like the cloud. Our revolutionary approach allows our users to build a global network without any hardware reliance or complex hop-by-hop configurations. There’s no need to download instances, manage software, or overbuild for maximum capacity. Instead, we provide a complete solution that’s AI and automation-ready today.

Many enterprises are adopting hybrid and multi-cloud architectures. How does Alkira’s platform handle the complexities of this hybrid model, particularly regarding security and operational efficiency?

In the age of cloud and AI, managing complex networks across diverse environments, public, private, and on-premises—has become a significant challenge for businesses. Forget the headaches of new hardware, software downloads, or wrestling with intricate cloud architectures. Alkira’s NIaaS addresses the “spaghetti plate” of hybrid cloud networking, by providing a single, consolidated view and control over your entire network.

This isn’t just about basic connectivity; it’s about agility. Our on-demand network and security services extend beyond the cloud, encompassing all business connectivity and your network backbone. Imagine establishing secure site connections or spinning up new backbone segments in minutes, not days. Services like load balancing and WAN optimization become intuitive, integrated features that scale automatically with your needs.

Alkira delivers a future-ready network that is simple, secure, and scalable. It empowers enterprises to fully harness the potential of multi-cloud environments and modern connectivity, fostering innovation, expanding market access, and sharpening their competitive edge in an AI-driven, digital world.

With the growing demand for AI workloads, Alkira’s platform aims to integrate cloud networks seamlessly. How does Alkira’s network architecture specifically support AI workloads, and what advantages does it bring to enterprises adopting AI at scale?

Traditional networks often fall short for AI workloads, being too complex, inefficient, and expensive. They struggle to scale with the fluctuating demands of AI training and inferencing, often pushing businesses towards heavy reliance on hyperscalers and raising data sovereignty concerns.

Alkira offers users a unified global network fabric that simplifies connecting on-premises and cloud resources. That makes building and managing AI-ready infrastructure much easier. Our platform allows for dynamic scaling of network resources, so businesses can seamlessly adjust to growing or changing AI workloads.

Alkira’s infrastructure supports both on-premises and cloud environments, providing the flexibility to design AI infrastructure that fits specific needs, including those looking to avoid sole reliance on hyperscalers. Our unique approach simplifies infrastructure management with a unified, scalable network backbone, freeing enterprises to focus on innovation. It also ensures optimal performance and cost by utilizing resources on-demand for both model training and inferencing. With a single control console, network operators gain full visibility and control across all environments, enabling fine-tuning of their AI workloads.

Can you explain how Alkira’s infrastructure overcomes the typical bottlenecks in AI cloud adoption, especially regarding performance and security?

The growing demand for specialized AI clouds, optimized for GPU-intensive tasks, presents a major challenge for enterprises integrating them into existing hybrid multi-cloud architectures. Traditional networking methods for connecting to AI clouds often involve complex manual setups, security inconsistencies, scalability bottlenecks, and fragmented management. So, organizations need a networking solution that seamlessly connects to AI clouds without compromising performance or security. Alkira’s NIaaS provides exactly that, enabling businesses to extend their hybrid multi-cloud networks into AI clouds with agility, security, simplicity, and scale.

Alkira’s baked-in agility allows rapid deployment through an intuitive point-and-click interface, connecting on-premises systems, public clouds, and AI cloud providers via Alkira’s global Cloud Exchange Points (CXPs). This eliminates hardware dependencies and accelerates AI workload deployment. For security, Alkira offers end-to-end encryption, advanced segmentation, and Zero Trust Network Access (ZTNA) natively, with integration for leading security vendors, ensuring consistent policies across all environments.

Simplicity is achieved through a unified management console, streamlining design, provisioning, and control, offering comprehensive visibility into network performance. Finally, Alkira provides elastic, on-demand scalability with its consumption-based NIaaS model, allowing businesses to expand connectivity as AI workloads grow without capacity limits or upfront hardware costs, ensuring low-latency, high-throughput networking. This cost-efficient pay-as-you-go model optimizes resource utilization and eliminates CapEx, supporting scalable AI initiatives without financial strain.

As AI adoption increases, scaling network infrastructure can be a challenge. How does Alkira’s platform ensure enterprises can scale efficiently while optimizing cloud costs?

As AI becomes increasingly central to enterprise strategies, organizations must reimagine their network architectures to support these advanced workloads. Traditional approaches—characterized by static configurations, security silos, and manual processes—will increasingly become competitive disadvantages.

Alkira’s Network Infrastructure-as-a-Service platform provides a strategic foundation for organizations seeking to fully leverage AI cloud capabilities without the complexity and limitations of conventional networking approaches. By delivering agility, security, simplicity, and scale within a unified framework, Alkira enables faster innovation, stronger protection, and more efficient operations.

For enterprises navigating the complex terrain of multicloud and AI integration, Alkira offers not just a networking solution, but a strategic enabler for digital transformation. The organizations that adopt this approach will be well-positioned to harness the full potential of AI while maintaining the governance, security, and operational efficiency required in today’s complex digital ecosystem.

Alkira’s platform offers a unified, on-demand multi-cloud network with integrated security and segmentation. How do you envision this approach evolving as enterprises increasingly adopt AI and machine learning workloads that require high-throughput, low-latency networking?

As enterprises increasingly embrace demanding AI and machine learning workloads, Alkira’s unified, on-demand multi-cloud network, with its integrated security and segmentation, will become an even more critical component of AI infrastructure. Alkira will deepen its integration with specialized AI-optimized datacenters, like Cisco and Nvidia’s Secure AI Factory. Its extensive network of Cloud Exchange Points (CXPs) will ensure ultra-low latency and high performance for data transfer between diverse AI compute resources, crucial for distributed AI model training and real-time inferencing.

Alkira’s platform will seamlessly abstract complexity, allowing enterprises to effortlessly connect various AI clouds, on-premises AI clusters, and traditional cloud resources into a single, cohesive AI infrastructure. Dynamic resource allocation and a subscription-based model will enable on-demand network capacity hyper-scaling; precisely matching fluctuating AI workload demands without over-provisioning. Security capabilities will strengthen with granular segmentation and advanced threat detection tailored for AI data flows. Centralized management will simplify operations, allowing enterprises to focus on AI innovation. Alkira’s platform will transform into an intelligent, adaptable network backbone, unlocking the full potential of distributed, high-performance AI across any hybrid multi-cloud environment.

AI is rapidly transforming industries across the board. How do you see AI further influencing the networking landscape in the next five to ten years, especially in terms of automation and real-time decision-making?

The AI wave has well and truly arrived and is impacting every industry in several ways, raising the bar for speed and production in the enterprise. AIOps, which is the use of AI to automate and streamline IT operations tasks, is going to be the biggest use case for AI in the networking space in the next few years. When it comes to networking, every capability needs to be tightly integrated, and AIOps makes meaning out of all the disparate information floating around. This is going to be critical for anomaly detection, which can help enterprises quickly identify the root cause of an incident, so they can resolve it before there’s any damage to their network.

It’s important to remember that although AI will be a helpful tool for networking as we move forward, we’re still far from AI handling network provisioning on its own. Networks are mission-critical, so using AI to automate an entire network is still too risky, especially if you don’t have the right personnel to monitor it and understand all the real-time decisions it’s making. An enterprise can lose millions of dollars in an hour if they don’t know where an issue happens. AI models will need to be trained for a long time for organizations to feel comfortable giving it more network responsibility.

With AI workloads becoming more critical, what’s the next frontier for cloud networking solutions? How is Alkira positioning itself to address the next wave of AI-driven network needs?

Security and trust are significant hurdles for AI adoption, as enterprises worry about potential negative impacts if AI systems aren’t adequately secured. This lack of trust can hinder full AI utilization. Alkira addresses these concerns by embedding a zero-trust security model into its network fabric, ensuring secure access and data transmission across all connected environments. This is crucial for protecting AI workloads and data.

Alkira’s platform enables segmentation, allowing enterprises to isolate traffic between users, applications, and environments, thus securing and ensuring compliance for Large Language Model (LLM) workloads. Our solution provides fine-grained controls with agility across multi-cloud and hybrid environments, ideal for managing LLM data flows. We also support integrated security services like firewalls and intrusion detection systems, enhancing network security without separate deployments. Alkira’s centralized management simplifies enforcing security policies and meeting regulatory requirements across AI infrastructure. By minimizing operational overhead, Alkira lets enterprises focus on their core AI initiatives, fostering trust and encouraging adoption of AI.

Alkira has introduced Load Balancer-as-a-Service to simplify and scale load balancing across cloud and on-premises environments. Can you elaborate on how this service enhances operational efficiency and supports the dynamic needs of modern enterprises?

Our Load Balancer-as-a-Service (LBaaS) simplifies and scales traffic management, directly enhancing operational efficiency for modern enterprises. It resolves the complexities of multi-cloud and hybrid load balancer deployments by integrating them directly into our Cloud Exchange Points (CXPs). This unique placement ensures optimal connectivity and eliminates the need for complex networking configurations.

Businesses can deploy load balancing with a simple point-and-click interface, making advanced cloud networking accessible to all skill levels while retaining full access to advanced features like SSL offloading and layer 4-7 traffic management. This flexibility supports balancing traffic across different cloud regions, multi-cloud scenarios, or integrating with existing hybrid infrastructure.

Benefits include rapid global deployment, efficient infrastructure-centric integration, and automated lifecycle management that centralizes provisioning and scales instances while maintaining policy control. Enhanced operational visibility through Alkira’s unified portal provides comprehensive insights. Our service supports both external and internal load balancing for multi-region, multi-cloud, and hybrid scenarios through a single, integrated platform, streamlining application delivery.

 Thank you for the great interview, readers who wish to learn more should visit Alkira.