Mercor Raises $750 Million in Series D Funding To Train AI On Every Job

OpenAI secured $11 billion in funding last week, catapulting its valuation to $1.25 trillion. However, a lesser-known player, Mercor, has now raised $750 million in its Series D funding round, pushing its valuation to an impressive $20 billion. This marks a doubling of Mercor’s value from just over a year ago. The startup, based in San Francisco, is on a mission to train AI for every job by compensating real humans to instruct AI on performing real tasks.

Founded three years ago by Brendan Foody, Adarsh Hiremath, and Surya Midha, Mercor initially focused on AI recruitment solutions. However, the founders soon realized that human experts are crucial for creating and testing AI models across various professions, including lawyers, doctors, bankers, and engineers. Today, Mercor boasts a network of over 5 million experts in 800 occupations across the United States, generating an annualized revenue run rate of more than $2 billion, as of July 2026.

The new capital will be utilized to train AI models using company-specific data, ensuring that the models perform effectively within the unique workflows of each business. In recent months, Mercor acquired Deeptune, a team that spent two years rebuilding hundreds of enterprise apps before integrating it with AI agents. These agents enable users to practice with realistic enterprise app copies without compromising actual systems.

As businesses continue to adopt AI agents for workflow automation, investors are doubling down on Mercor's potential, seizing the growing market opportunity for training AI on all tasks.