Snorkel AI has raised $350 million in a Series E round at a $3.5 billion valuation. Insight Partners and S32 led the financing, while Addition, Lightspeed, Greylock, GV, and Wells Fargo also participated. The valuation is nearly triple the $1.3 billion valuation from the company’s Series D 17 months earlier.
Snorkel initially provided software for automating data labeling before shifting last year to completed datasets through a data-as-a-service model. Its hybrid approach combines software and models that generate data synthetically with subject matter experts. The company serves AI labs and corporations building training datasets and simulated environments.
Snorkel says its annualized revenue run-rate has reached $375 million, representing an 18-fold increase over the last 12 months. The company attributes this growth to AI labs’ demand for high-end training data. Other AI data companies have also reported rapid growth, including Mercor, Handshake, and Micro1.
Mercor’s gross annualized revenue has reached $2 billion, Handshake has hit $1 billion, and Micro1 has scaled to $500 million. These companies pay roughly 60% to 70% of top-line income to domain specialists, so their net annual revenue is substantially lower than their headline gross figures. Snorkel sells complete datasets and reinforcement learning environments , recording payments to human experts as costs of goods sold.
Snorkel launched commercially in 2019 after four years of research by co-founder and CEO Alex Ratner and his team at a Stanford AI lab. The company’s current offering reflects its shift from data-labeling automation toward completed datasets and simulated environments.