The human story of an AI startup

In the small city of Cooch Behar in Bengal, a world away from start-up hotspots like California’s Sand Hill Road or Bangalore’s HSR Layout, Tapan Barman grew up watching his mother run orphanages.

Today, Barman is the co-founder and CEO of Mihup, a more versatile, multifaceted and distinctly Indian version of Alexa or Siri, with options in vernacular languages.

Founded in 2016, Mihup received Rs 45 crore in seed funding from Accel Partners. A further Rs 12.5 crore came in 2019 from Accel and Ideaspring Capital. And in December they received Rs 11 crore from Accel, Ideaspring, Core91 VC and YuMe Networks founder and CEO Jayant Kadambi. The company’s clients include Tata Motors, Swiggy, Panasonic, Hyundai and the Indian National Congress.

And yes, for their first three years, Mihup did move from Kolkata to Bangalore.

On a video call from Kolkata on December 29, Barman speaks about his inspirations. “The biggest force behind me becoming an entrepreneur was my mother,” he says. “She was an enterprising woman who ran orphan homes in our city. I was in a central government job initially. It was the most comfortable kind of thing for any middle class family, right? But I was not happy. I wanted to do something of my own. My mother gave me enough strength to follow that path. Sadly, she passed away last year.”

Another human relationship was the catalyst for the next step. Barman teamed up with his childhood friend from next door, Biplab Chakraborty, now the COO at Mihup. The third member of the founding team, Sandipan Chattopadhyay, came on board through a common friend.

Tapan Barman

“I have known Biplab for 30 years, we went to the same school and college,” says Barman. “We were looking for funds and we reached out to Sandipan through a friend. We told him the concept and what we wanted to build. He liked it. That’s how the Mihup journey started.”

Barman’s advice to aspiring entrepreneurs is to first get the team right. It was his biggest challenge. He quotes from the Jim Collins book, ‘From Good to Great: Why Some Companies Make the Leap, and Others Don’t’.

“The most important thing is the team. If you refer the Jim Collins quote, it is ‘First Who, then What’,” says Barman. “Unless you have people with almost missionary zeal, you won’t be able to execute your idea. For us, finding talent in Kolkata was not easy. When you are building a start-up, speed is key. It’s ok if you make some mistakes. Skill is important too, but you can learn that along the journey. But you need great attitude to fill those gaps.”

Great attitude is one of the reasons Barman and the almost 60 staffers of Mihup, spread around India, work 11 hours a day – from 9 am to 8pm. But that’s just the official timing. Barman rarely switches off, despite having a young family. Meditation and reading are his only small escapes.

“The major problem [in Kolkata] is with attitude,” he says. “Even today, it is hard for bright people to join start-ups. Most of the talent will go to a big company that offers a handsome package.”

Barman believes, however, that the success of Bengal start-ups like Wow Momo and Teabox will inspire youngsters from the state to become entrepreneurs.

His other tip for those wanting to start a business was to think big. It is why he admires Elon Musk. “He made what seemed impossible possible,” Barman says. “Also, you have to visualise what will happen after 10 years. That will shape whatever you are building today.”

Talking about the role of investors in the company, Barman says they were “lucky” to have support from people like Accel’s Subrata Mitra. There was no constant pressure from them, nor bureaucratic demands.

“We had our first board meeting one whole year after their investment. Throughout the year we never met [with Accel]. They allowed us to focus on our work and did not divert our focus on board meetings and such,” says Barman. “Now I know why Flipkart became so big. Subrata was the person who wrote the first cheque to Sachin and Binny (Bansal). You need that kind of investor who will support you, not dictate what should be done.”

Barman reiterates the assurance that AI will not replace humans. Asked if he could change the much-feared term ‘Artificial Intelligence’ to something else, he says, “Even before AI, we were empowered with some form of data. Because we decide a lot of things based on data. With the help of AI, we are getting data with some recommendations. So we could call AI ‘Alternative Intelligence’. But ultimately humans will make the decisions.”