Organizing human intelligence for the entire economy

The next decade of AI will be defined by how companies bring increasingly capable models into every part of their business.

AI is only as useful as the human expertise behind it, and as models grow more capable, the demand for specialized, domain-specific knowledge compounds. Every job in the economy will need its own agents that can perform tasks reliably and efficiently.

We spent the last two years building the infrastructure to train and evaluate frontier models on real-world work with human intelligence. We’re now bringing those same capabilities to enterprises across the economy to evaluate what those agents can do, what they cost, and how to improve their performance.

A growing number of companies are realizing that owning their intelligence is essential. Starting from a capable open model, enterprises can post-train with a fraction of the resources required at the frontier. The playbook is already visible across the industry: build public evals, post-train a model to cut costs, then train models for your customers. Model builders and compute providers supply the foundation. We give every enterprise the data and evaluation needed to make those models their own.

The capabilities Mercor has built for the frontier labs are now being adopted by big tech, developers of open models, application layer companies, and the Fortune 500.

Working directly with frontier labs and enterprises gives us a distinct advantage. Both draw on the same expert network, evaluation methods, and post-training infrastructure. Labs want models that do economically valuable work, and the clearest picture of that work lives inside enterprises. What we learn on one side makes us better on the other.

Building for every profession

We’ve created a new category of work with more than five million experts in our network getting paid to train frontier AI models based on their unique knowledge.

In the United States alone, there are more than 800 distinct occupations, each containing workflows that models have yet to be evaluated against. Professional work depends on two kinds of knowledge that no model inherently has: how a task gets done and how a specific company does it according to its systems, policies, or quality bar. Without both, an agent fails in production.

We work with enterprises and experts in their fields to build custom evals based on the company’s actual workflows. The results show how models perform against the company’s requirements and help determine where additional training or practice in realistic environments could improve performance.

That is why we acquired Deeptune earlier this year. Its team spent two years recreating hundreds of enterprise applications where AI agents can practice professional workflows.

Investing in our next chapter

Today, we're announcing our $750 million Series D valuing Mercor at $20 billion. The round included an employee tender along with participation from numerous investors, including Benchmark, Felicis, Fundomo, funds managed by affiliates of Apollo, G Squared, General Catalyst, Glade Brook Capital, NVIDIA, Salesforce, Sapphire Ventures, StepStone Group, XTX Ventures, and Zoom.

This year, our team has grown beyond San Francisco. We’ve opened offices in New York and London to be closer to the enterprises and labs we serve and recruit from some of the strongest talent pools in the world.

This latest funding accelerates our investment in the infrastructure and research for evals and post-training and in building more environments where agents can learn complex workflows.

We’re seeing the impact of AI investment in our business. We already spend 3x as much on inference as on employee salaries, and it pays for itself by making our team better at their jobs. If that ratio holds as AI spreads through the rest of the economy, it points to something closer to double-digit GDP growth over the next five years, nearly 3x the rate we’re used to.

More people will work alongside AI agents that can perform increasingly complex tasks, creating new opportunities for companies to apply human expertise at scale. We’re organizing the human intelligence required to help enterprises measure, train, and improve those systems as they become part of the entire economy.