Mercor organizes human expertise to power the AI economy, working as a marketplace for domain experts and a supplier of human data for AI labs and enterprises. Experts such as physicians, lawyers, engineers and consultants find contract work through Mercor, which cites an average contracted rate of 109 dollars per hour. Companies use the same network for human data and evaluations, benchmarks and reinforcement-learning environments built on real professional work, and for enterprise AI projects that move from pilot to production.
Two groups use it. Professionals with specialist knowledge, from clinicians to software engineers, browse listings such as expert survey roles or software engineer positions paying 100 to 150 dollars per hour. On the buyer side, AI researchers at frontier labs need human evaluation at scale, and enterprise leaders want to apply AI to their own workflows with help from forward deployed engineers. Businesses sitting on workflow data can also explore its data monetization program.
Concrete scenarios include a lab commissioning expert graders to assess model output in medicine or law, a team using the APEX benchmark family to test agents on long-horizon professional tasks in accounting or software engineering, and an insurer mapping underwriting workflows before deploying production AI. The data monetization offering masks more than 60 categories of sensitive identifiers automatically and leaves the organization in control of what is shared. Recent news on the page includes an agreement to acquire Deeptune to build training environments across more industries and roles.
