doola.

A company that lives inside the chat where companies now begin

Somewhere right now, a founder in Lagos, or Lahore, or Lisbon is typing a sentence into ChatGPT: "form an LLC using doola." Thirty seconds of setup, a company name, a state, a Stripe checkout - and a real United States company exists. No lawyer in the room. No notarized stack of forms. No flight to Delaware.

That is doola in 2026: the only company-formation platform sitting natively inside Claude, ChatGPT and Replit, quietly turning the most intimidating part of entrepreneurship into a line of conversation. It started as a website. It is becoming infrastructure.

The U.S. economy had a velvet rope

Here is an inconvenient truth about the world's largest market: getting into it was designed for people already in it. To form a U.S. business as a non-resident, you needed an address you didn't have, a tax ID that required a Social Security number you couldn't get, a bank that wouldn't return your emails, and an accountant who charged like one.

The talent was already global. A developer in Karachi could ship the same software as one in San Francisco. But the developer in Karachi couldn't get paid like one - because the plumbing of incorporation, banking and compliance was bolted shut from the outside. The gap wasn't ability. It was access.

Leave Dropbox. Bet on the do'ers.

Arjun Mahadevan studied at Wharton, learned the trade on Dropbox's data and growth teams, and then did the thing growth people are supposed to talk founders out of: he left a good job to start a hard company. At 25. The company began in 2020 as StartPack - itself a pivot from a startup called Standups - before becoming doola.

The bet, made with co-founder and CTO JP Pincheira, was deceptively simple. If you could compress formation, an EIN, U.S. banking access, bookkeeping, taxes and compliance into one place - a "Business-in-a-Box" - then geography would stop being destiny. Y Combinator agreed. So did Nexus Venture Partners and, later, HubSpot Ventures.

Fun fact: the company you now form inside an AI chat was, five years ago, a scrappy MVP chasing its first paying customer. The first customer said yes. The rest is a lot of paperwork - handled.

One box, the whole back office

The pitch is that you never see the plumbing. Behind a single dashboard, doola handles the unglamorous machinery that turns an idea into a compliant, bankable company - and an AI Co-Founder that increasingly does the clicking for you.

Translation: the boring half of running a company, outsourced to software that doesn't take weekends.

The numbers do the arguing

Mission statements are cheap. Reach is not. doola's case rests on how far the velvet rope has actually moved - measured in founders, countries and reviews left by people who had nothing to gain from being kind.

Then there are the partners. In 2026, doola wired formation directly into the tools founders already live in: Claude and Replit first, then ChatGPT's 800M+ weekly users, then Lovable for no-code builders and Vercel for developers shipping straight to production. The launch post alone drew 225,000+ organic impressions in a single week.

A billion founders, eventually

doola says it wants to help one billion founders start and scale their dream business. It's the kind of number that sounds like a pitch deck until you remember the size of the problem: most of the planet's would-be entrepreneurs have never had a clean on-ramp to the world's deepest market.

The mission isn't charity. It's arbitrage on a truth the company keeps repeating - that talent is distributed evenly and opportunity is not. Close that gap and you don't just do good; you build a very large business doing it.

The on-ramp is disappearing - on purpose

For most of business history, the barrier to starting a company wasn't the idea. It was everything around the idea: the entity, the bank, the books, the filings, the fear of doing it wrong. doola's bet for the next decade is that all of that collapses into software - and increasingly, into a conversation with an AI agent that just handles it.

If that bet lands, the interesting question stops being "how do I form a company?" and becomes "what should the company do?" Which is the only question that was ever worth a founder's time.

So return to that founder in Lagos, mid-sentence in a chat window. Five years ago, the U.S. economy would have answered with a form, a fee and a polite no. Today it answers with a confirmation email. The velvet rope is still there. doola just keeps quietly cutting it.