In this interview, Jonathan Kader, VP of Product and Analytics at Quvia, shares his path into product leadership and how his team is using AI and ML to solve real-world customer challenges—like optimizing connectivity for end-users, automating network management and turning analytics into action. From natural language interfaces to task-savvy AI agents, he offers a look at how Quvia is driving innovation and redefining what’s possible in intelligent network orchestration.
From Meta to Knotel to APT, you’ve worked across a range of tech product and analytics roles. How have those experiences influenced the way you lead product teams now?
Jonathan Kader: Throughout my career, I’ve consistently worked at the intersection of data, product and enterprise customers—helping businesses translate complexity into clarity. Each role has reinforced the importance of pairing technical rigor with deep customer empathy, especially when serving large or sophisticated clients.
At Meta, I worked with global MNOs and ISPs, often co-developing solutions that required precision at scale. At Knotel, I helped design tools for enterprise clients navigating flex office needs. And at APT, I worked on advancing decision analytics software that enabled Fortune 500 companies to use experimentation and data science in their everyday operations.
These experiences shaped how I lead product today. When you're building for enterprises, the key isn’t just functionality—it’s adaptability. You need to serve the 80% that’s consistent across your customer base while designing for the 20% that’s highly contextual. That means not just thinking in terms of feature sets, but also deployment realities, internal stakeholder needs and integration requirements.
I have also seen the importance of solving for both the end user and the economic buyer - whose priorities are often in conflict. The product must drive daily efficiency for users while delivering strategic ROI to leadership. Balancing intuitive UX with business impact has been a throughline of my work—and it’s central to how I build teams, roadmap features and prioritize investments.
What motivated you to join Quvia, and how does its mission align with your own vision for product innovation?
JK: I’ve always been drawn to companies at inflection points—where product innovation can shift an entire industry, not just optimize a small corner of it. Quvia was clearly at one of those points when I joined, and still is today.
After talking with the team—including some former colleagues from Meta—I saw not only the scale of the opportunity, but also the clarity of the vision. Our founder and CEO Benny Retnamony’s ambition for the company was bold but grounded, and aligned closely with how I think about product: as a lever for transformation, not just incremental improvement.
What made Quvia stand out was the combination of technical depth, market need and team alignment. I knew this was a place where I could help build meaningful, zero-to-one products—not just manage feature velocity. That’s what I was seeking: a place where the work we do in product and analytics can move markets, not just metrics.
At Quvia, you lead both product strategy and customer-facing analytics. How do you ensure those two areas stay tightly aligned to deliver value to customers?
JK: We think of customer analytics not just as a service function, but as a prototyping engine for product innovation.
Our core platform delivers significant value in network orchestration and visibility. But because we generate rich, high-context data, many of our customers come to us with adjacent use cases—questions about refunds, network trends or operational optimization that aren’t part of the core product... yet.
The customer analytics team works closely with these customers to rapidly build tailored dashboards, models and insights using the data we already produce. It’s consultative, but also deeply strategic. When we see repeatable demand across customers or verticals, we evaluate whether these insights should be formalized into product features. This cycle—from prototype to product—is one of the reasons we’ve been able to move fast and stay aligned with real-world customer needs.
This approach comes from my experience at Meta and APT, where customer facing data science teams often led early-stage innovation before handing it off to engineering for scale. It’s a playbook I believe in: validate through data and partnership, then productize with confidence.
Ultimately, the tight alignment between product and analytics ensures that we don’t just build what we think is valuable—we build what’s already being used, loved and demanded.
Can you highlight a recent product initiative or feature that best reflects Quvia’s approach to solving customer pain points?
JK: One of the clearest examples is our recent work on supporting consumption-based plans within our Network Capacity Controller (NCC) product.
The connectivity landscape is evolving rapidly, particularly with the rise of LEO satellites and hybrid networks. Historically, bandwidth in the satellite industry was sold using fixed CIR/MIR models. But today, we're seeing a shift—especially from newer LEO and 5G providers—toward consumption-based pricing. This creates real operational complexity for our customers.
Previously, operators tried to manage usage manually—restricting certain traffic types or setting conservative daily limits to avoid overages. But these methods were blunt instruments. They relied on guesswork and frequently resulted in either overspending or degraded user experience.
What we built at Quvia changes that. Working closely with customers and iterating through our algorithms and data science team, we developed a dynamic system that intelligently allocates traffic across consumption-based plans while preserving a high quality of experience (QoE) for end users. Our platform can now anticipate bandwidth needs, make real-time adjustments and optimize against both cost and performance—something no other product in the market could do with this level of sophistication.
This initiative perfectly reflects our DNA: deeply listening to customer pain points, applying technical depth to solve novel problems, and moving quickly from insight to solution. Because of our architecture and talent, we can tackle challenges others shy away from—and deliver value that’s both immediate and defensible.
AI and automation are reshaping the satellite and telecoms industries. How is Quvia using AI/ML today, and what opportunities are you most excited about in this space?
JK: AI and ML aren’t new buzzwords for us—they’ve been foundational to our platform since day one. Long before generative AI entered the spotlight, we were using machine learning to automate network decisioning and optimize for QoE. That’s been central to how we deliver value since Quvia was founded in 2019.
Take our Consumption-Based Management (CBM) feature as an example. Behind the scenes, it’s driven by a suite of machine learning models that continuously analyze usage patterns, predict bandwidth demand and allocate resources across the month—all while ensuring SLAs are met. It’s a complex optimization problem, and AI is what makes it tractable in real time.
But we’re also incredibly excited about the opportunity with generative AI. Our Q product—an LLM-powered interface—allows customers to query their data and generate custom dashboards using natural language. No more submitting product feedback to us or navigating complex BI tools—just ask and receive. This is a major unlock for non-technical users and speeds up decision-making across the board.
And we’re just scratching the surface. We believe AI agents will play a growing role in how networks are managed. Because Quvia sits in a uniquely contextual position—between networks, customers, and third-party providers—we’re exploring agent-based automation that can proactively create tickets, coordinate with external systems or even resolve issues autonomously. In the future, we envision agent-to-agent communication becoming a core mechanism of modern network operations.
It’s an exciting time. We’re not just reacting to the AI wave—we’re helping define what the next generation of intelligent infrastructure looks like.
How do you strike a balance between delivering sophisticated analytics and maintaining accessibility for non-technical users?
JK: The key is empathy and proximity to our customers. We never build in a vacuum. Our product sits at the intersection of highly technical network orchestration and diverse customer roles—ranging from seasoned network engineers to business stakeholders. It’s our responsibility to bridge that gap.
A great example is our approach to QoE metrics. Historically, the industry leaned on raw network data—packet loss, latency, queue depth, DNS resolution times—metrics that, while useful, don’t always map directly to end-user satisfaction. Non-technical users would often hear conflicting signals: the network looked “healthy,” yet passengers or crew were frustrated with performance.
We reimagined that experience by developing a proprietary suite of QoE scores that translate technical telemetry into a single, understandable measure of user experience. Now, an airline executive can quickly answer: Was video streaming possible at 720p on this flight? Was a corporate VPN usable at that time?—without digging through graphs or logs.
Importantly, we don’t lose the fidelity of the underlying data. Technical users can still access and interpret raw metrics, and we’re transparent as we can be about how the QoE scores are derived. But we’ve found that simplifying the signal for non-technical users is a force multiplier. It improves trust, speeds up decision-making and ensures our product becomes a shared language across teams.
That’s the philosophy: maintain technical rigor, but lead with clarity.
Looking ahead, what trends or shifts do you believe will shape the next generation of data-driven products—and how is Quvia preparing for them?
JK: There’s no question that generative AI is the defining trend—and it’s already reshaping how we think about product design, especially in the data and network infrastructure space.
Traditionally, customers relied on dashboards—static interfaces built around predefined questions. But as LLMs become more capable, we’re seeing a shift: users no longer want to search for insights—they want to converse with them. That’s the future: natural language interfaces that sit on top of complex systems and return answers, not just visualizations.
At Quvia, we’ve beta launched our Q product to support that vision. With Q, if QoE drops on a ship, our customers don’t need to dig through multiple dashboards or cross-reference RF stats, throughput levels and weather data. They can simply ask, “Why is QoE dropping on Ship A?”—and get a structured, explanatory response in seconds.
That’s not just time-saving—it’s transformative. What used to take 15 minutes now takes 15 seconds. And as we layer on AI agents, this gets even more powerful: imagine a system that doesn’t just diagnose issues but initiates fixes—reprioritizing traffic, filing provider tickets, or dynamically adjusting bandwidth allocation.
This is where Quvia is headed: from insight to action, powered by AI. The future of data-driven products isn’t just analytics—it’s autonomous intelligence. And we’re building the foundation now.
