Vihaar Nandigala and the art of finding the right door

A restaurant has a kitchen. Someone else needs a kitchen. Between those two facts, Vihaar Nandigala found a business. KitchenKonnect matched unused cooking space with people who wanted to use it, giving restaurants another way to earn revenue. The idea had the pleasing clarity of a good seating arrangement: put the right people together and something useful might happen.

Making the introductions was another matter. Nandigala went to restaurants and knocked on doors. Before there was a conversation, there was the work of deciding where to go. Before a pitch, there was a list. Years later, as cofounder and CEO of Orange Slice, he is still occupied with that preliminary question: who actually needs what you are selling?

It is a small question with an expensive habit of being answered badly. A persuasive message delivered to an indifferent recipient remains an interruption. Nandigala's career has brought him from Michigan to J.P. Morgan to a San Francisco software company, but the restaurant doorway provides a useful place to begin. There, the difference between a plausible customer and an interested one was close enough to knock on.

A kitchen, a list, and an exit

Nandigala grew up in Michigan with immigrant parents and went to the University of Michigan. His early interest in making things came through mobile games. Around sixteen, he began experimenting with building one himself, learning to code along the way. Playing the game was apparently the easier half of the arrangement.

KitchenKonnect brought a different education. He built it as a solo founder, without outside funding, drawing on contractors for occasional work. Eventually, a restaurant chain that had been a customer bought its customer lists. He describes selling the business at nineteen. The useful detail is the buyer's proximity: someone already using the service saw value in the demand it had gathered.

His account of leaving is unusually practical. Growth had stalled, the work required considerable time, and he wanted room to tackle another problem. An early exit can sound like a final examination passed with distinction. His telling leaves space for the more useful lesson: a founder can learn a great deal from a business and still decide someone else should run it.

The work before the sale

He estimates that sales and go-to-market work consumed roughly eighty percent of his time in the restaurant business. Some of that meant pitching. Much of it meant preparation: scraping Google Maps, examining reviews, choosing restaurants and planning which ones to approach. The administrative prelude had acquired a leading role.

Later, he and his future cofounder spent about a year operating an agency that helped other businesses find suitable customers. They tried existing sales tools and assembled workflows for clients. That experience supplied a second view of the same difficulty: knowing what research would help did not automatically make the research easy to carry out.

It also gave the eventual product a demanding first customer in its founders. They wanted a tool that could turn their own ideas about finding buyers into working processes. The distance between an idea and its execution became part of the brief. A sales representative should be able to ask a useful question without first acquiring a second profession in connecting software.

A cofounder before another idea

Nandigala and Kishan Sripada met sophomore year on a Bollywood dance team at Michigan. Sripada became his college roommate and, eventually, Orange Slice's cofounder and CTO. A dance team is an entertaining place to find a business partner. It also gives people an opportunity to work together before anybody starts discussing equity.

In his account of choosing what came next, Nandigala puts trust early in the sequence. After running a company alone, he wanted a cofounder before settling on the next idea. That ordering gives the dance-team anecdote some substance. The friendship preceded the corporate biography; the company inherited a relationship that had already existed elsewhere.

Sripada brought his own experience of building, including choreography software Formi and engineering work at Ramp. Nandigala had worked at J.P. Morgan. Their shared company grew from different encounters with the practical work of reaching customers. There is something agreeable about a sales-software partnership with a rehearsal history. At least the first exercise in coordination did not require a calendar invitation labelled “alignment.”

The spreadsheet gets a research department

Orange Slice entered Y Combinator's Summer 2025 batch. Its current product description gives the spreadsheet a new job: each column can be TypeScript, with AI writing the code from a request in chat. A user describes the information needed; the software generates columns and runs the work across rows. The familiar grid becomes a way to conduct research.

The company lists more than one hundred built-in enrichments in its SDK, covering contact information, company data, web scraping and other research. Columns execute according to their dependencies. That last detail matters in an ordinary way: one answer may be needed before another question can sensibly be asked. Automation still has to get the order right.

The website offers concrete uses: identifying companies through funding announcements, researching hiring, filling gaps in customer records, and monitoring competitors. It also describes columns that send information into a CRM or trigger an outreach step. The attraction is continuity. Research should remain useful when it moves from an interesting finding to someone's actual working day.

For Nandigala, the connection to the restaurant business is easy to see. The setting has changed, and the tools are more elaborate. The underlying task still involves finding evidence that a particular business might need something. A spreadsheet cannot guarantee a welcome. It can give its user a better reason to introduce themselves.

The scale of a list can be misleading. Ten thousand names can look productive long before anyone has established why those people belong together. A smaller set of well-understood prospects asks more of the researcher and less of the recipient. That trade gives the product a clear test: does the information help a person make a sensible next move? The answer has to survive contact with an actual customer, who is under no obligation to admire the machinery.

A seed round, then the small print of building

In February 2026, Orange Slice announced $5.3 million in seed funding. The round was co-led by 1984 Ventures and Moxxie Ventures, with Paul Graham participating as an angel investor. Nandigala said the money would support engineering hiring. The financing gave the founders resources to continue turning their sales experience into a product other people could use.

The company's engineering writing also reveals the kind of work behind the chat interface. A March 2026 essay argues for code as the basis of automation, with smaller agents handling focused tasks. It makes a case for branching logic, reusable functions, testing and version control. Those are less photogenic subjects than a fundraising announcement. They become rather attractive when a workflow breaks halfway through.

That concern with dependable execution has a smaller, more personal counterpart in Nandigala's own updates. In one post, he described three broken links: a billing button that reached an error, an account shortcut that could go to the wrong screen, and a public link to a page that had never been built. The problems shared a cause. Each link constructed its destination separately.

The fix was to use one shared place for those addresses and test it. His conclusion was grounded in the customer's experience: someone clicking a useless button sees a broken product. It is an unglamorous observation, which is part of its appeal. A company can have a theory about the future of sales and still owe someone a functioning billing page.

The last sentence is his

Perhaps the clearest account of how Nandigala wants to use AI comes from his own prospecting routine. Each morning, an AI system sends him relevant conversations: LinkedIn discussions, Reddit complaints about competing tools, people asking for something like Orange Slice. The machine gathers the context and points out where an answer might be useful.

Then he opens the posts, reads them, and writes the replies himself. His objection to automated messages is recognizable to anyone whose inbox has developed a suspicious abundance of polished enthusiasm. Messages can sound interchangeable. Research, by contrast, helps him arrive with a reason to speak.

“Then I write the message.”Vihaar Nandigala, on his AI-assisted prospecting routine

There is a pleasing restraint in that division of labour. The founder selling automation reserves a manual task at the end of his own process. He uses software to improve his attention, then spends that attention personally. The restaurant doors have become online conversations. He still wants to know which one to approach, and he still intends to do the talking.