Behind Sense AI: Vijay Umapathy, VP of Product

Vijay is a product management leader who joined Contentsquare via the Heap acquisition and has been with the company for just over a year. Prior to that, he led product initiatives at Heap and worked on a wide range of products, including advertising, Inbox by Gmail, and consumer robotics at Jibo. Vijay is especially drawn to new technologies and zero-to-one products—those that carve out value in emerging spaces full of uncertainty and opportunity.

We sat down with Vijay to learn more about his journey in product management, his work on Sense, and his perspective on the future of AI. Here’s what he shared:

What sets Sense apart from other AI tools in the digital analytics space?

I think what stands out most about Sense is the power and diversity of its capabilities. We didn’t want to create a point solution that only works for one type of user; we set out to build a true intelligence and approachability layer across the entire product.

When we began working on Sense, we conducted extensive foundational research. It became clear that the biggest challenge, both in our product and across the analytics space, is that analytics can feel overwhelming for novices. Learning the terminology, understanding which types of analysis to use, and interpreting charts can all be intimidating.

In our industry, there’s a lot of “demoware”—tools that answer one or two types of questions but don’t actually teach users anything. I’m proud that Sense does the opposite: it educates users in context, helping them learn as they work.

In fact, a customer told me just this week, “It teaches you as you go.” That kind of feedback reminds us we’re on the right path, and we’re only getting started.

As a product leader, what role did you play in shaping Sense?

In the beginning, my role was all about building alignment and pitching our executive stakeholders on why we should invest in generative AI– this was before it became the top priority for every SaaS company. Once we secured a few resources and could demo early AI features, the product started speaking for itself.

Our initial AI team at Heap was small: two engineers, half of a designer, and me as the PM. As we developed our first capability, a natural language interface to product analytics, we also worked to build internal buy-in. Beyond executive meetings, we created momentum through weekly Loom videos that showcased our prototypes evolving in real time. These videos didn’t just show outcomes—they captured the process, which helped turn skeptics into supporters.

That internal traction ultimately led to a 5x increase in AI investment across the company since last summer. It also evolved into what’s now the AI Accelerator at Contentsquare, where I lead product management across four dedicated AI teams. One of the most meaningful parts of this journey has been helping bring together the talented engineers, designers, and PMs who are building the AI capabilities you see in Contentsquare today.

How has being closely involved in developing Sense helped you grow in your role as a product leader?

I think a big part of building a career in product, and in tech more broadly, is making smart bets. Going all-in on AI in 2023 was one of those. It meant working with limited resources early on, but it allowed us to build crucial expertise that quickly became valuable to the company and its leadership. It also gave me the opportunity to shape our AI strategy, even while only implementing small pieces of it at the time. Now, with greater resources, we’re able to move much faster and bring that vision to life at scale.

How does Sense simplify the work for teams who may not be experts in data or analytics?

Sense removes much of the complexity that typically comes with analytics. Instead of needing to know which analysis to run, how to configure it, or how to interpret the results, users can simply ask a question and Sense responds in plain language, complete with data interpretation.

It’s not limited to English, either. As of now, Sense supports all of Contentsquare’s languages, making it accessible to a global user base.

Beyond answering questions, Sense also actively navigates the Contentsquare interface for you. This helps users move smoothly across workflows, bypassing any friction that might come with switching between tools or views.

How did your team approach making the technology both powerful and easy to use?

Large language models (LLMs) are incredibly powerful, but they can also feel like a black box. From the start, we knew we had to take a deeply customer-oriented approach to building our AI features. That meant prototyping early and testing often with real users, real questions, and real data.

We had our first working prototype just four weeks after the team was formed, and those early tests taught us a lot. For instance, we discovered that many users wanted to understand why the AI chose a particular type of analysis or configuration. In response, we added a button labeled “Explain your reasoning,” which has been a big hit and a key driver of transparency and trust.

Ongoing customer research, both during prototyping and post-launch, is also central to our development process. No AI feature is ever truly “finished”; there’s always more we can do to make it clearer, more intuitive, and more user-friendly.

A big thank you to Vijay for sharing his journey, insights, and the vision behind Sense. His leadership and thoughtful approach to product development continue to push the boundaries of what’s possible with AI at Contentsquare.