- Mercor raised a $750M Series D at a $20B valuation, double its last round.
- NVIDIA, Salesforce, Benchmark and General Catalyst are among the investors.
- It is pushing into enterprise AI evaluation against Scale AI and Surge AI.
San Francisco-based Mercor, an AI data company that helps organisations train and evaluate AI using human expertise, has raised $750 million in Series D funding at a $20 billion valuation.
The round includes participation from 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.
The funding will accelerate Mercor’s research, evaluation and post-training infrastructure, including realistic environments where agents can learn professional workflows. The aim is to help companies measure and set standards for how reliably AI agents complete tasks and at what cost.
The raise follows Mercor’s $350 million Series C at a $10 billion valuation in October 2025, doubling its valuation in about a year. In July, Mercor announced an annualised revenue run rate above $2 billion. Tech Funding News reported at the time that this is gross billings, with contractors keeping roughly 60% to 70% of it, which put net revenue at an estimated $600 million to $800 million.
From AI training data to enterprise intelligence
Founded in 2023 by Brendan Foody, Adarsh Hiremath and Surya Midha, Mercor initially connected AI companies with professionals providing specialised knowledge to train advanced models.
The company says its network includes more than five million experts and serves frontier AI labs, technology companies, AI startups and Fortune 500 businesses. TFN’s July report described that figure as candidates vetted, with about 30,000 contractors in its rotating pool. Mercor now has 700 employees.
It is expanding beyond supplying training data towards helping enterprises build AI systems that understand their own operations.
“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 Foody, co-founder and co-CEO.
Mercor spent the past two years building that infrastructure alongside frontier AI labs, and is now taking it to enterprises across industries.
Why enterprises need customised AI evaluations
AI models can perform well on public benchmarks and still fail when navigating company-specific software, policies or operational exceptions. Mercor addresses this with customised evaluations, or evals, that measure AI performance against an organisation’s own standards.
“Every company deploying AI needs to know whether it works, and the answer is different for each one,” said Hiremath, co-founder and co-CEO.
The company builds benchmarks from actual enterprise workflows to show where additional post-training is needed. Its approach builds on APEX, its public benchmark series measuring AI performance on economically valuable professional work. Earlier this year, Mercor acquired Deeptune, whose team spent two years recreating hundreds of enterprise applications as training environments where agents can practise complex workflows before deployment.
“Mercor has built something hard to replicate… We’re thrilled to partner with Brendan, Adarsh, and team to help them scale in the Enterprise market,” said Jai Das, president, partner and co-founder at Sapphire Ventures.
Das added that frontier labs have so far been the main buyers of expert-driven training data, and that enterprises will now want to evaluate and train agents on their own work.
Mercor faces competition from Scale AI and Surge AI
Mercor operates in a growing market for specialised AI training data, human feedback and evaluation. Scale AI is a major competitor, having raised $1 billion in Series F funding at a valuation of nearly $14 billion in 2024. Meta later invested $14.3 billion for a 49% stake.
Surge AI, another rival in expert-driven training data, was reportedly weighing its first outside raise at a valuation above $25 billion. Snorkel AI raised $350 million at a $3.5 billion valuation as it shifted from data-labelling software to expert-built datasets and simulated environments. Meanwhile, micro1 raised $35 million at a $500 million valuation in September 2025, led by 01 Advisors, to expand its expert network.
Mercor’s differentiation lies in combining its expert network with customised benchmarks and realistic training environments, which positions it beyond traditional data labelling.
Mercor believes the next phase of AI adoption will come from applying existing models to professional work, not only from building new foundation models.
The challenge will be scaling its enterprise business against established AI data providers. For Mercor, the opportunity is to become the layer that helps businesses deploy AI agents reliably.
