AfterQuery raises $30M Series A at $300M valuation for expert-curated AI training data
The AMW Read
Incremental update to a data-infrastructure player with notable revenue claims; segment-level significance due to validation of expert-curated data as a distinct market within AI training infrastructure.
AfterQuery raises $30M Series A at $300M valuation for expert-curated AI training data
AfterQuery, a provider of specialized AI training data, announced a $30 million Series A round led by Altos Ventures, with participation from The Raine Group and existing backers Y Combinator, BoxGroup, and Latitude Capital. The company, which maintains a vetted network of roughly 100,000 domain experts, claims it has surpassed a $100 million revenue run rate. Its data offerings include supervised fine-tuning pairs with chain-of-thought reasoning traces, reinforcement learning reward signals based on expert-designed rubrics, and "Computer Use Trajectories" datasets that capture human software interaction for training AI agents.
Why it matters: AfterQuery exemplifies the emerging "human-in-the-loop data-as-a-service" pattern, where specialist tacit knowledge — the unspoken reasoning of financial analysts, senior engineers, and other professionals — is systematically extracted and structured for model training. This directly addresses a growing recognition that public web text and synthetic data alone are insufficient for frontier-level reasoning, especially in high-stakes verticals. The company's rapid revenue ramp to $100M run rate at a $300M valuation (a 3x revenue multiple) signals that enterprise buyers are willing to pay premium prices for verified expert data, potentially creating a new moat based on access to scarce human judgment rather than compute scale.
Ground expert take: AfterQuery sits at the intersection of two structural forces: the capital-compression arc pushing foundation-model labs to seek differentiated training data, and the agent-economy demand for high-fidelity behavior traces. While the $30M round itself is modest by AI infrastructure standards, the implied $100M run rate suggests a lean, high-margin business that could become a critical supplier to the entire model ecosystem. The key open question is scalability: whether AfterQuery's expert curation pipeline can maintain quality as it grows, and whether competitors can replicate the network effect of 100,000 verified specialists. If successful, the company may shift the data-IP debate from copyright disputes over scraped content toward a market for licensed expert cognition.
#AfterQuery #TrainingData #AIInfrastructure #ExpertData #ChainOfThought #AgentTraining
