First, congratulations on your selection as a 2025 BigDATAwire Person to Watch! Back in 2012, you and your eventual Atlan co-founder, Varun Banka, were building a big data platform for prime minister of India. Did you ever think that work you were doing at SocialCops would lead to a successful company?

Absolutely not – and yet, looking back, it feels almost inevitable. At the time, we weren’t optimizing for success. We were optimizing for impact. We didn’t set out to build a company – we set out to solve meaningful, high-stakes problems.

From counting buildings with satellite imagery to converging 600+ messy data sources, SocialCops gave us a front-row seat to some of the most painful, chaotic, and manual data challenges in the world. And when you live through that pain long enough, you either quit – or you build something better. Atlan was born out of that “enough is enough” moment.

We weren’t trying to build a startup. We were just obsessed with solving the problem the right way.

Atlan has become one of the top data catalog providers over the past few years, and was the far and away leader in the most recent Forrester Wave for Enterprise Data Catalogs. What do you attribute that success to?

Our biggest competitive advantage is care.

At Atlan, we operate with a core principle: customers > company > team > me. That hierarchy shapes every decision, every line of code, every roadmap debate. We truly care – about solving real problems, about making our customers heroes in their organizations, about being a real partner in their journey.

This level of empathy has helped us build trust. It’s why we’ve consistently been the top-rated solution across industries and customer review platforms. It’s also why we’ve been able to innovate ahead of the curve.

We were the first to launch Atlan AI. The first to operationalize Data Mesh and Data Products in a catalog. We pioneered Active Metadata and redefined the category – not as a documentation tool, but as a living, breathing fabric of the modern data stack.

We didn’t just talk about “shifting left.” We built workflows that integrate metadata natively inside engineering tools. Every one of those bets came from listening deeply and caring intensely.

And that care will be our edge going forward. As our customers face the biggest shift of their careers in this new AI-native world, they won’t need just another vendor. They’ll need a partner they can trust. We plan to show up with the same level of care, empathy, and innovation they’ve always known us for.

Data governance is hard. What is the one most important thing that practitioners do to improve their odds of success, or at least minimize the pain?

Start with the business problem. Not the technology.

After working with 200+ data teams, we’ve built something we call the Atlan Way – a set of hard-won lessons about what actually makes governance succeed. Not just the tech, but the people, the program, and the operating model.

Most governance programs fail for one of three reasons:

1. They never get up and running.
   The metadata stays dry. Implementation is too manual. It’s too hard to maintain. That’s why we built Atlan to be automation-first and to shift left – deeply integrating into the data producer workflow. Governance shouldn’t be a one-time setup. It should be a sustainable, long-term habit – part of how you build and ship data products every day.
2. They never get adopted.
   This is where our change management philosophy kicks in: don’t force it. Take technology to your users – don’t bring your users to the technology. That’s why Atlan shows up where your team already works: inside Slack, Microsoft Teams, BI tools, and data warehouses. We meet people where they are, not where we wish they’d be.
3. They’re not future-ready.
   Change is the only constant in the data ecosystem. Two years ago, nobody was talking about vector databases. Last year, they were everywhere. This year, the conversation has already moved on. Governance systems can’t be brittle. That’s why we’re building a fully open platform – so governance doesn’t slow teams down, it sets them free.

At the end of the day, we believe governance should be invisible. It shouldn’t feel like control. It should feel like enablement. Embedded in the workflow. Built for real humans. And always evolving.
 

Atlan’s strategy is to serve as the metadata control plane, sitting above the data tool stack to govern data via metadata. That’s not how data practitioners are accustomed to doing everything within their tool. What is the secret to changing those old habits?

The secret is simple: you don’t change behavior—you design around it. 

One of our earliest lessons at SocialCops was that people revert to what’s easiest. You can’t brute-force new workflows. So instead of trying to fight that, we built Atlan to be the connective tissue – not a new silo. Our philosophy is to meet people where they are, not where we wish they were.

That’s where Active Metadata comes in. Most metadata platforms act like passive libraries – great for documentation, but disconnected from real work. We flipped that mode. Atlan activates metadata across the stack – embedding it into tools teams already use: GitHub, Slack, Teams, dbt, BI tools, and data warehouses.

We’ve brought metadata into engineering workflows, where producers actually build and ship data products. We’ve helped data consumers find trusted context right inside the tools they already use. This is what we mean by shifting governance left – governance that feels like a feature, not a friction.

Because at the end of the day, “Metadata isn’t a layer you add. It’s the foundation you build on.”

GenAI tools and LLMs are proliferating in enterprise data stacks. What difficulties do these new tools and technologies pose to data governance?

We’re no longer in a digital-native world. We’re entering an AI-native one.

The most interesting thing about LLMs is that they now understand language – but they don’t understand meaning. Only humans can teach that. And as LLMs start doing more of the work humans once did, one question matters most: can you trust it?

Can you trust the data that trained the model? Can you trust the model that produced the output? Can you trust the AI-generated action that affects your business, your customers, or your brand?

That’s where governance steps in. Not as policy enforcement, but as a system for context and trust.

In the AI-native enterprise, governance isn’t a back-office function. It’s a frontline enabler. The companies that move fast and build trust will be the ones that win. But that’s only possible if governance evolves into an intelligent, embedded, real-time capability.

We believe this is governance’s leapfrog moment – a chance to move from being a cost center to a competitive advantage. As businesses rewire their products and processes with GenAI, the real question won’t be “Can we do this?” It will be “Can we trust this?”

That trust has to be systemic. It can’t stop at the data. It has to flow through the entire lifecycle of decisions, models, and automation. That’s the role of Active Metadata as a semantic layer: making meaning machine-readable, making governance invisible, and helping AI act with context and care.

And that’s why “In the AI-native era, governance isn’t a blocker. It’s the unlock.”

What can you tell us about yourself outside of the professional sphere – unique hobbies, favorite places, etc.? Is there anything about you that your colleagues might be surprised to learn?

I’m the only Prukalpa in the world – literally. My parents say they thought of SEO before Google existed, and honestly… they weren’t wrong.