Five months. That’s how long it took AfterQuery to go from a $300 million valuation to $3.2 billion. The AI training-data startup’s latest round, first reported by Forbes, marks more than a 10x jump since its $30 million Series A in April — and according to Y Combinator partner Gustaf Alströmer, it’s the fastest any startup in the accelerator’s history has gone from launch to unicorn status.
The founders, now 22 and 23 years old, were in Y Combinator’s Winter 2025 cohort just 18 months ago. That trajectory would be remarkable in any sector, but it’s especially striking in the crowded training-data space. AfterQuery has positioned itself against incumbents like Mercor and Scale by focusing not on answer accuracy but on something more ambitious: encoding how professionals actually work. Doctors, lawyers, and other specialists don’t just verify responses — they demonstrate the decision patterns and reasoning models need to complete tasks the way experts would.
Revenue tells part of the story. The San Francisco company reported a $100 million annualized run rate in April and counts Nvidia, Legora, and Korean AI lab Motif Technologies among its customers. That’s real commercial traction, not just investor enthusiasm. But the valuation math deserves scrutiny. A 10x jump in five months implies AfterQuery’s growth has accelerated dramatically — or that investors are paying a premium for a scarce asset in a market where high-quality human expertise for model training is becoming the bottleneck.
The broader context matters here. As frontier labs push toward agentic AI systems that don’t just answer questions but execute multi-step tasks, the demand for training data that captures professional workflows — not just correct answers — has exploded. AfterQuery’s bet is that teaching models to think like a radiologist or negotiate like a corporate lawyer is worth more than teaching them to recite facts. If the valuation signals anything, it’s that investors believe the next phase of AI development hinges on exactly that kind of behavioral data. Whether the company can scale its specialist network fast enough to justify the price remains the open question.
