In this edition of The Exchange, I sat down with Naveen Tewari, founder and CEO of InMobi Group and a former McKinsey colleague I’ve known for many years.
Naveen is a pioneer of India’s start-up ecosystem, having built the country’s first tech unicorn. Today, he leads a global consumer platform that connects more than 235 million people daily through InMobi and its subsidiary, Glance AI, creating real-time, AI-powered content and commerce.
We spoke about the early days of InMobi, how its teams are releasing products in just 48 to 72 hours, the rise of AI and why 80 percent of all coding could be done by AI by the end of this year, and what it means to build and lead in a rapidly changing world. Our conversation was wide ranging—from product decisions and leadership shifts to a shared appreciation for both cricket and excellence.
For anyone navigating transformation and exponential change, Naveen’s journey is a remarkable master class in clarity, conviction, and moving fast—especially when the path is still unfolding.
Building boldly: From vision to global impact
Naveen, I think what you’ve accomplished is phenomenal—India’s first unicorn. Let’s go back to the beginning and start with the vision. When you started, what did you want to do?
Going back almost 18 years, after working at McKinsey and finishing business school, I realized one thing: I should try to do something on my own. The idea that somebody coming from middle-class India could start a company in India was not a well-accepted concept.
The vision at the beginning was simply to start a company; there wasn’t necessarily a construct of what this could really become. I thought if I could just get something off the ground, we would have achieved everything. Not even in my wildest dreams did I expect things to become as big as they have.
The first construct was very hard: to build a technology product from India for the world. We had all seen the success of services organizations in the late 2000s.
We made three defining choices:
- We chose to build a technology product over services, which was unconventional for India back then.
- Within a few months of starting the company, we decided to build a global platform; without the advantage of a Silicon Valley tag, this wasn’t easy.
- We chose to hire the best engineers, preferably those who might have otherwise left India. We gave them an opportunity to build a global platform by selling them the vision. It was often difficult to convince their parents, because the idea of joining a start-up and working for an unknown company is risky—the parents, of course, wanted a good life for their children.
These decisions weren’t immediately validated, but they were pivotal. Being able to build a product and have the world realize that India could build one that was globally accepted—built by engineers who felt proud to work for a smaller company—proved successful not just in our own company, but across India.
AI will code the future, and it’s happening now
Naveen, fast-forward to today and the world of AI. You mentioned 70 to 80 percent of code will be generated by AI. What does that mean for companies and talent?
Coding is going to get automated; it’s a given. We’re seeing engineering teams adopt the latest AI tools to essentially reinvent themselves.
People who are using these tools—and letting AI automate 70 to 80 percent of the easier part of the job—will find that the last 20 percent is where expertise is needed. That means the remaining work gets done faster, creating more time to take on more work that requires expertise. That’s where automation is headed.
For example, integration and infrastructure optimization are getting done faster, so you may need to be more of a full-stack engineer.
I absolutely believe we’re headed toward a world and workforce where increased automation will reshape job roles. The engineer role has already changed. But job continuity is still there—if you actually learn to use these tools. Job continuity basically means you will do a lot more work than you did before. We’re seeing this in our own company.
At InMobi, teams have adopted this and are releasing products within 48- to 72-hour cycles, driven by AI-assisted workflows. Prototyping costs are approaching zero. The old rule of spending weeks planning before executing is being flipped. Now we build, test, learn, and iterate faster.
The duration cycle is so rapid that instead of putting all the thought in ahead of time, you convert it into rapid iterations. This changing approach to software development is also changing the role of a full-stack engineer, who can move from product manager to engineering manager to coder, quality controller, and software deployer—all in one.
From personalization to the personal internet
How is AI changing the consumer experience, especially in digital advertising? And in that context, what type of innovations have you introduced?
This is a very fascinating space. Every industry is undergoing a change. In advertising, for example, the internet could become more beautiful.
The experience of an ad from a complete user perspective is going to be much better. Performance advertising, which is at the core of digital advertising, will also become more effective. Everything will be optimized: Models will determine the right ad experience for individuals in their environment. Ads will be generated in real time, focused on the right product, experience, and context. AI will auto-optimize everything, knowing the exact ROI that needs to be delivered. Every element of this will be automated—essentially converting it into machinery that scales itself across the internet.
For example, if you’re reading an article on CNN, the content could be dynamically rewritten or regenerated to retain its original meaning while seamlessly embedding a new, personalized ad experience. Similarly, if you’re watching a video, the video itself could be regenerated to incorporate relevant advertising without altering its core message.
We’re moving into a world of real-time content generation for everyone—it’s going to be more personal. Advertising will become personal as content becomes personal.
In retail, we just launched our product, Glance AI. We think retail globally will change as well. Glance AI is fundamentally doing AI commerce, essentially bringing a product and the consumer together using generative AI in a way that’s never been done before. With personal content generation, advertising will follow that same path.
Privacy, trust, and the role of technology
Naveen, you referenced this idea of personalization, which has been a big theme here. From a societal standpoint, there are many concerns about privacy, as well as differences by geography—the US versus Europe versus India or China. As you build a global platform, how do you start to think about privacy in the context of sensitivities across different parts of the world?
Yes, it’s a very critical topic. Personalization—or leveraging consumer data to create something—is at the core, because in most of these situations, consumers don’t fully realize what’s happening with their data. So systems, organizations, and platforms have to ensure they’re not misusing that information.
The systems are getting to a place where policymakers have set up tight guardrails, such as GDPR [General Data Protection Regulation] in the European Union and the European Economic Area. California has its own policy. Organizations need to make sure there’s a tight balance between what services they can give to consumers versus what they cannot. There will always be this tussle between the consumer experience that can be delivered and what organizations can leverage or not leverage.
The world is actually moving beyond just personalization to personal internet. With generative AI, one of the biggest changes happening is not just that content gets rearranged in a way that’s more suitable to what you want, but that it actually gets regenerated—in a way that’s only applicable to you. So you’re moving quite significantly toward a much more advanced personal internet experience where generative AI plays a big role.
On one hand, you have to take care of the privacy challenges, but the technologies that are really coming out are making things much more personalized and personal. Therefore, I believe that to protect consumers’ personal data, you need a technical solution to this problem, and that solution has to be one where the data does not leave the premises of the consumer.
The models are the ones that get trained and then come back. So you’re essentially going to move to a world where the training of the model is what happens across different elements, but the data stays with you on your own premises, whether it’s on your handset or your device. Therefore, a technical solution is the better solution.
Leading through fog: What AI-forward means
One of the discussions we’ve been having is ‘What does an AI-forward company look like?’ If you start to rebuild a company to be AI forward, what does that look like? What will the next three to four years in your planning horizon look like? Can you even start to predict?
The beauty of this situation is that it is so hard to predict. For the first time, we have no idea how to predict anything.
You could sit there and talk about the inability to predict and then not do something—that’s the worst place to be. The second is to say, “OK, I really don’t know what’s going to happen, but I’m going to move forward.”
That’s where most organizations need to sit. They should ask, “Do I have an exact view of what an advertising stack would look like in the world of AI? No, I don’t.” Companies should move forward and build rapidly because the cost of building has decreased. It’s like entering a fog with limited visibility—you won’t see the full path up front, but it becomes clearer as you move forward. Given the rapid pace of change, I wouldn’t recommend moving slowly. Act decisively. Advance rapidly.
We have a vision with Glance AI to show the world how commerce can change. There will be a new way to shop, and we’re building it. It took us a grand total of 12 months to build, and we launched it recently. I can’t say this is the only way shopping will be disrupted, but we’ve made a bold claim and committed to it. What’s certain is that everything is going to change—and we’re taking on two of the biggest industries out there: advertising and commerce.
I believe an AI-forward company is one that takes a bold stance by committing to build and trust models. You have to believe that these models will learn, evolve, and start doing meaningful things. And the beauty of this approach, the third and most difficult element, is learning to let go. That’s not easy.
The future lies in system design. But you don’t design the system—you design the guardrails. The system designs itself.
When we launched Glance AI, we didn’t script every detail. We simply set the boundaries. The models then took over, iterated, and created. We then fine-tuned the outcomes.
Excellence, cricket, and finding your 0.1 percent
Naveen, let’s turn to two other topics, which I know have been close to both of us growing up. Here in my room, I actually have a cricket bat—I know cricket is a passion of yours too. The other topic that I think has been a passion of yours is excellence. At McKinsey, we call it distinctiveness.
Let’s talk about both of those topics: cricket and excellence—and how they shaped you growing up.
This question really hits close to home, especially since the Royal Challengers Bangalore just won their first-ever Indian Premier League title. I wasn’t there, but it was craziness. You could hear everybody celebrating.
I am very passionate about cricket. I wanted to be a cricketer; it was my childhood dream, and I worked hard toward this. While I am pretty good at cricket, l realized that I was not in the top 0.1 percent needed to make it professionally.
You have to identify your area of excellence, your natural strength and natural talent, and put in an insane level of hard work there so you can become the 0.1 percent. That’s where distinctiveness comes from. It’s all about aligning your effort with your natural edge.
Final thoughts
Naveen Tewari’s journey is a powerful reminder that the future belongs to those bold enough to challenge norms and fast enough to act before they’re certain.
From building India’s first unicorn to leading AI innovation in mobile engagement, Naveen embodies what it means to be an AI-forward founder. He doesn’t wait for clarity; he creates it through motion.
Whether it’s letting models shape outcomes, building guardrails instead of road maps, or redefining what it means to code, Naveen is reframing what leadership looks like in a world transformed by AI. He doesn’t wait for the fog to lift. He moves into it fast and lets the system evolve.
For leaders navigating the AI era, there’s a lesson here: You don’t need full visibility to build the future. You need a clear point of view and the courage to act on it, striving to be the 0.1 percent.
Naveen Tewari is the founder and CEO of InMobi Group, a global leader in mobile advertising and consumer technology. Under his leadership, InMobi became India’s first unicorn and now reaches more than 235 million daily users via Glance AI, its AI-powered lock-screen platform. A former McKinsey consultant and Harvard Business School alumnus, Naveen has been recognized with the Ernst & Young Entrepreneur of the Year award and featured in Fortune’s 40 Under 40. He is also an active start-up mentor and investor, committed to building India’s digital future.
Asutosh Padhi, senior partner and global leader of firm strategy, is responsible for driving the strategic vision, accelerating the firm’s pace of innovation, and strengthening the partnership model for the next century. He was previously the North America managing partner, leading the firm across the United States and Canada, and was a member of the Shareholders Council, the firm’s equivalent of a board of directors.
He is also the coauthor of The Titanium Economy, a book that explores the industrial tech sector and the bright future that it can help create.
Comments and opinions expressed by interviewees are their own and do not represent or reflect the opinions, policies, or positions of McKinsey & Company or have its endorsement.
This interview was recorded on June 4, 2025.
This piece was originally posted on LinkedIn.com on July 7, 2025, as part of Asutosh Padhi’s interview series, The Exchange.