Mercor Secures $750m Series D Funding Round at $20bn Valuation, Doubling Value in Six Months

Mercor Doubles Valuation to $20bn in Six Months

Mercor, the San Francisco-based AI training startup, has raised $750 million in a Series D funding round that values the company at $20 billion1

1

2

2

. The Mercor valuation represents a doubling from the $10 billion valuation the company achieved when it raised $350 million in October 20253

3

. The rapid appreciation reflects investor enthusiasm for AI infrastructure companies that bridge the gap between frontier AI models and real-world business applications.

Major Tech Players Join Series D Funding Round

The Series D funding round attracted participation from Benchmark, Felicis, Fundomo, General Catalyst, Nvidia, Salesforce, Sapphire Ventures, and Zoom1

1

. Funds managed by affiliates of Apollo, G Squared, Glade Brook Capital, StepStone Group, and XTX Ventures also joined the round2

2

. The funding round included an employee tender offer, though Mercor did not disclose how much of the $750 million went to the tender1

1

. Mercor is backed by Jack Dorsey and venture capitalist Peter Thiel3

3

.

From AI-Powered Hiring Platform to Training AI Agents

Mercor began as an AI-powered hiring platform but has transformed into a network of more than five million subject-matter experts who are paid to train and evaluate AI models with their professional knowledge1

1

2

2

. The company's main buyers so far have been frontier AI labs, but the new funding is meant to sell the same service to companies across all industries. "Every job in the economy will need its own AI agents, and every company will need a way to train and evaluate them on its own work," said Brendan Foody, co-founder and co-CEO1

1

.

Custom AI Agents Built for Real Business Workflows

Mercor builds custom tests, known as evals, from each company's real workflows and uses the results to train and evaluate AI models on that company's specific work, often starting from an open model1

1

. Deeptune, acquired earlier this year, recreated hundreds of enterprise applications as practice environments for AI agents1

1

. "A model can score well on a public benchmark and still fail on the systems, policies, and edge cases that define how a specific business runs," said Adarsh Hiremath, Mercor's other co-founder and co-CEO1

1

. This approach addresses the gap between general-purpose AI models and the specialized needs of individual businesses.

Inference Costs Already Exceed Employee Salaries

Brendan Foody revealed that Mercor already spends three times as much on inference costs, the cost of running AI models, as it does on employee salaries1

1

. If that ratio holds across the economy, Foody argued, it points to something close to double-digit GDP growth over the next five years1

1

. This metric signals a fundamental shift in how companies allocate resources, with AI infrastructure potentially becoming a larger cost center than human labor for certain operations.

AI Startups Capture Majority of Venture Capital

According to Crunchbase, AI-driven startups raised $102 billion, or 64% of global venture capital, in the third quarter of 20262

2

3

3

. This investor enthusiasm has fueled a sharp rise in private funding this year, creating a growing group of heavily capitalized startups carrying multibillion-dollar valuations2

2

. Mercor's rapid valuation growth exemplifies this trend, as investors bet on companies that can help businesses deploy custom AI agents for professional tasks2

2

. The company also opened offices in New York and London this year1

1

.