What sparked the idea for Clientell while you were still at BITS Pilani?

While at BITS Pilani, I worked with multiple startups and eventually moved to Bangalore in my third year. I was dabbling in various businesses, trying to understand how different SaaS companies operated. A short stint in San Francisco helped me realize how bloated many business processes had become—dozens of tools, but minimal efficiency.

Why are we paying people to maintain tools when AI could manage or replace them entirely?

This question gave birth to Clientell - a platform that brings automation and intelligence to revenue operations. Instead of relying on humans to update CRMs or generate reports, our AI agents do it seamlessly, freeing up teams to solve more strategic problems.

What were some unexpected challenges you faced in your early startup journey?

Fundraising wasn’t as big a hurdle for me - I already had a good network, thanks to my internships and consulting gigs during college. But recruiting talent was a significant challenge.

I had never held a traditional job, so I didn’t fully grasp how to pitch my vision to someone with more experience. Convincing top-tier talent to leave stable roles and join a young startup takes more than just enthusiasm - you need clarity, confidence, and patience. It taught me a lot about leadership and people management.

What made your partnership with your co-founder, Neil Sarkar, click?

Neil and I were college roommates - we practically lived and built side projects together. From selling goods online to building a semi-commerce platform in college, we were always experimenting. That history gave us a shared language and rhythm.

What I’ve really learned from Neil is discipline. In college, we’d wake up at noon and code all night. Running a startup requires a completely different mindset - one where structure, pace, and long-term planning become crucial. Having someone who helps you evolve into that mindset is invaluable.

What advice would you give to students looking to enter the startup world?

Don’t wait for the “big idea.” Just start building. More importantly, talk to real customers. The best insights don’t come from hackathons or pitch decks - they come from listening to the pain points of actual businesses.

For instance, when we first started consulting with companies, I offered to fix their CRM problems for $2,000 a month. Several said yes. That gave us immediate validation and revenue. Over time, we realized we could automate these fixes with AI. That’s how the real product emerged. Chase the problem, not the hype.

What’s one myth about startups that you wish more people understood?

You need a big team to build something big. That’s simply not true.

One of our goals with Clientell is to see how far a small team, just 10 to 15 brilliant people, can take a company. AI is a force multiplier, and if used wisely, it can help small teams do the work of hundreds. That’s where we’re heading: towards leaner, smarter startups.

AI is evolving rapidly. How do you keep up and ensure your products stay ahead?

AI doesn’t work like traditional software. There’s no fixed output - you’re always working with probabilities. For example, if you ask an AI agent to write an essay, one person might love the result, and another might not. So, we developed over 100 internal parameters to evaluate our agents: accuracy, consistency, latency, adaptability, and so on.

In one real-world case, we deployed Clientell for a company whose sales pipeline had completely stalled. The AI restructured their CRM, reprioritized stale leads, and helped them recover over $50,000 in potential revenue.

But beyond code, the hardest part is understanding the user’s intent. What are they trying to do? What friction can we remove for them? Solving those questions is the real challenge - and the fun.

Where do you see your work going in the next few years?

We’re now focused on building fully autonomous AI systems—agents that not only take instructions but decide on their own what to prioritize, when to act, and how to deliver value.

We’re also experimenting with how to scale this efficiently, without needing large teams. We want to build an AI Lab, potentially in Bangalore or San Francisco, that pushes the boundaries of automation and product design with just a few highly skilled builders.