Christina Qi left behind a hedge fund trading $7 billion a day for a farm in Utah. Her new startup just raised $97 million to rival Bloomberg

Christina Qi spent a decade making money for billionaires,  building Domeyard LP—a high-frequency trading hedge fund that traded up to $7.1 billion a day. Then, she shut the whole thing down to start Databento, a market data infrastructure company.

Databento’s pitch is almost comically unsexy compared to her last venture: sell financial data feeds to the same banks and hedge funds she’d left. The startup just raised a $97 million Series B led by NEA, Fortune learned exclusively, with participation from DRW Venture Capital, Redpoint Ventures, and Tribe Capital. The round drew over $300 million in demand, the company said. And Qi is running it all from a farm in Utah.

With only 24 employees, Databento is already profitable—a combination so unusual that, by Qi’s own admission, her investors keep telling her to just spend more money.

“We haven’t really touched much of that $97 million,” she told Fortune. “Our investors are telling us, spend, spend money, and we’re trying. We’re buying servers, buying stuff like crazy, and yet we’re still profitable every month.”

The raise brings Databento’s total disclosed funding to around $127 million. No valuation has been made public, but the capital is already earmarked: the company plans to expand from its current servers housed directly inside stock exchanges to more than 20 data centers worldwide, and has secured an additional 100-plus petabytes of storage—more than doubling its previous footprint.

The problem Databento is solving has bothered anyone who has ever tried to get institutional-grade market data. Qi recalls sending more than 100 emails over 11 months just to procure data from the largest provider on the planet (Qi wouldn’t say directly, but I can: Bloomberg). Sample data arrived on a thumb drive, via snail mail. Databento’s answer is a service where users can add Microsoft stock data to a cart like an e-commerce order and pay only for what they use—a deliberate contrast to Bloomberg Terminals running $20,000 to $27,000 per seat per year and LSEG’s Refinitiv at comparable rates. Meanwhile, global financial market data spending hit $49.2 billion in 2025, and it’s still growing.

Databento’s edge is partially technical. The company uses specialized chips to capture and clean raw price data (every single buy and sell order, in real-time) from 80-plus trading venues worldwide. It is the only vendor delivering that full picture over a regular internet connection. But the real lever is reach: 400 to 500 people sign up daily, ranging from UC Berkeley students vibe coding trading strategies in their dorm rooms to the Abu Dhabi Investment Authority. 

“90% of revenue does come from the large financial institutions and AI labs,” Qi said. “But the beauty is we’re able to attract the best of both worlds.” That includes Nvidia, which Qi confirmed is both a paying customer and technology partner. OpenAI is also a customer.

Databento’s greatest risk is the one that’s taken down most other scrappy data startups: getting acquired before they could become generational companies. In 2022, MayStreet sold to LSEG, and last year BML was acquired at Series B for around $250 million—exactly where Databento is now. “We’re going to bring this way past a billion,” she said. “This is just the starting line.”

This story was originally featured on Fortune.com