Snorkel AI Raises $350M at $3.5B Valuation on Data Boom

Data startup triples valuation in four months

Snorkel AI has secured $350 million in new funding at a $3.5 billion valuation, CEO Alex Ratner told Reuters, reflecting surging demand from frontier AI laboratories for sophisticated training data and simulated environments. The round, led by Insight Partners and S32 with participation from Addition, Greylock, and Wells Fargo, values the San Francisco company at nearly three times its $1.3 billion valuation from a $100 million raise in May 2025.

The valuation jump tracks explosive revenue growth. Snorkel's annualized revenue run-rate has reached $350 million, up from roughly $20 million a year earlier, driven primarily by a data-as-a-service business launched in September 2025.

Why it matters

Snorkel's growth illustrates a fundamental shift in AI development economics. As models grow more capable, the bottleneck has moved from compute to high-quality, domain-specific training data that requires expert human judgment. Companies that can deliver finished datasets at scale—rather than just labeling tools—are capturing outsized value as frontier labs race to train increasingly sophisticated systems. The rapid valuation increase also signals investor conviction that data infrastructure will remain critical even as AI capabilities advance.

From software to finished data products

Founded in 2019 by researchers from Stanford's AI lab, Snorkel initially sold software for data labeling. The company has since pivoted to supplying finished datasets and reinforcement learning environments directly to customers, a shift that reflects the evolving needs of AI developers moving beyond basic annotation work.

Ratner explained that demand has grown as AI labs seek harder, higher-stakes data to train and evaluate increasingly capable systems. Snorkel now operates what it calls an "agentic data development platform" that combines human experts with thousands of specialized AI models and agents to create and validate data. Human specialists design scenarios, tasks, and evaluation criteria, while AI systems automate much of the quality assurance process.

The company maintains a network of tens of thousands of specialists across fields including coding, law, and medicine. Rather than charging for human labor hours, Snorkel sells the data products these experts generate—a model Ratner said allows the company to pay specialists more generously while maintaining margins.

Human expertise remains essential

"Our strong view is that 100% of the data that labs will get value out of will have some human input in the foreseeable future," Ratner said. "But 100% of that data will have to use synthetic and automated approaches to keep up with this complexity."

Snorkel's customer base includes frontier AI labs, hyperscalers, enterprises, and the U.S. federal government. Coding data represents one of the company's largest demand areas.

Market momentum

Venture capital continues flowing into startups that supply frontier AI labs with human-annotated training data. The market transformed after Meta's $14.3 billion acquisition of a 49% stake in Scale AI in June 2025. Competitors including Mercor and Surge AI have also attracted investor interest amid strong revenue growth.

The details were first reported by Reuters.