AfterQuery: YC's Fastest Unicorn at $3.2B Valuation

An AI training-data startup founded by two former high-school friends has become the fastest company in Y Combinator’s history to reach unicorn status. On September 1, 2026, multiple outlets reported that AfterQuery closed a new funding round valuing the San Francisco company at $3.2 billion — roughly a tenfold jump from the $300 million valuation it carried just five months earlier, according to reporting from TechCrunch, Forbes, and Decrypt.

The speed is the story. AfterQuery went through Y Combinator’s Winter 2025 batch, and the company reached a billion-dollar-plus valuation faster than any YC startup before it, YC partner Gustaf Alströmer said in comments cited by the reporting. Co-founder Spencer Mateega, 23, is among a pair of founders in their early twenties now steering a company at the center of one of AI’s fastest-growing supply chains.

From a $30M Series A to a $3.2B round in five months

The trajectory is compressed even by 2026 standards. In April 2026, AfterQuery announced a $30 million Series A led by Altos Ventures at a $300 million valuation, with The Raine Group participating alongside existing backers Y Combinator and BoxGroup. At the time, the company said it had crossed a $100 million annualized revenue run rate — an unusually large revenue base for a company that had existed for barely a year.

Five months later, the valuation has climbed to $3.2 billion. Reporting characterized the new financing as a Series B; one source cited by Forbes said the company is already profitable and had lined up a lead investor for the round, though the identity of that lead was not disclosed in the initial reports. AfterQuery has not published a full investor list for the new round.

The pace invites comparison to the broader AI funding environment, where valuations for companies touching model development have compressed timelines that once ran for years into a matter of months. AfterQuery’s leap from seed-stage obscurity to a multibillion-dollar mark stands out because it did not come from building a consumer app or a foundation model — it came from selling the raw material those models are trained on.

What AfterQuery actually sells

AfterQuery describes itself as an applied data-solutions research lab. In practice, the company builds and curates the training data that AI labs use to improve their models — and it has positioned itself around the industry’s shift toward what it calls high-end human reasoning data.

That shift matters. Early large language models were trained largely on vast scrapes of public web text, then refined with techniques like instruction tuning and reinforcement learning from human feedback. As the frontier has moved toward models that reason through multi-step problems — math, code, science, and complex agentic tasks — the bottleneck has moved with it. Labs increasingly need expert-generated demonstrations and evaluations: worked solutions from people who actually understand the domain, structured so a model can learn the reasoning process rather than just the answer.

That is expensive, specialized work, and it sits squarely in the supervised and reinforcement-learning pipelines that turn a raw pretrained model into something useful. AfterQuery’s pitch is that it can supply this reasoning-grade data at scale to many of the largest labs at once — a position that turns it into a shared supplier for competitors who otherwise share very little.

Why the data layer is suddenly valuable

For most of the past decade, data labeling was treated as a low-margin, commoditized function — armies of contractors drawing bounding boxes and rating outputs. The economics of reasoning data are different. The people who can produce a correct, well-explained solution to a hard mathematics or programming problem are scarce, and the labs competing for their output are among the best-capitalized companies on earth.

That scarcity has pulled a handful of data companies into the spotlight. AfterQuery is riding the same wave that has lifted established players in the human-data market, but its growth rate and its youth make it the sharpest example yet of how much value has migrated to the data layer. When frontier labs are spending tens of billions on compute and chasing every incremental point of benchmark performance, the marginal dataset that unlocks better reasoning is worth paying up for.

The revenue figures reflect that. A $100 million-plus run rate reached inside roughly a year, paired with reported profitability, is the kind of financial profile that justifies a venture markup — even a tenfold one — in the eyes of investors betting that demand for reasoning data will keep compounding as models get more capable and more agentic.

The context: a frothy but selective AI market

AfterQuery’s raise lands in a market that has been both euphoric and discriminating. Capital is flowing freely to companies with a defensible position in the AI stack, while the bar for what counts as defensible keeps rising. The definition of a unicorn — a private company valued at $1 billion or more — has almost lost its novelty in AI, where nine- and ten-figure rounds have become routine and the more telling metric is how fast a company gets there.

At the same time, the biggest names in the sector are racing toward the public markets. Anthropic has moved toward an IPO and floated an enormous addressable-market pitch to investors, and rivals are queued behind it. AfterQuery is a very different kind of company — a supplier rather than a lab — but its valuation is a direct read on the same thesis: that spending on AI capability, and everything feeding it, has further to run.

What it means

AfterQuery’s ascent is a signal about where value is accruing inside the AI boom. The headline attention still goes to foundation-model labs and chipmakers, but the data layer underneath them has quietly become one of the most lucrative — and most contested — parts of the stack. A company that supplies reasoning-grade training data to many labs at once occupies a structurally strong position: its customers are locked in an arms race, and better data is one of the few levers that reliably improves model quality.

Who wins. AfterQuery’s founders and early backers, obviously, but also the broader class of human-data companies now being repriced upward. If reasoning data is the new bottleneck, the firms that can source it at scale hold pricing power over some of the wealthiest buyers in tech.

The risks. Concentration cuts both ways. A supplier whose revenue depends on a handful of frontier labs is exposed if those labs bring more data work in-house, if synthetic-data techniques improve enough to reduce demand for expensive human demonstrations, or if the capital cycle that funds lab spending turns. A $3.2 billion valuation reached in five months also sets a high bar: the company now has to grow into a number that assumes reasoning-data demand keeps compounding.

What to watch. Whether AfterQuery names its lead investor and formalizes the round; whether its reported profitability holds as it scales; and whether the human-data market as a whole can defend its margins against synthetic data and in-house labeling. For now, the takeaway is simple — in 2026, the fastest path to a unicorn ran not through building a model, but through feeding one.