What is the Core Business Model of Mercor?

Artificial intelligence is getting smarter every single week, yet the biggest secret in Silicon Valley is that the smartest machines still need human teachers. That single reality just turned a three-year-old startup called Mercor into a $20 billion powerhouse.

Founded in 2023 by three college dropouts—Brendan Foody, Adarsh Hiremath, and Surya Midha—Mercor recently closed a massive $750 million Series D funding round in November 2026, doubling its valuation from $10 billion in just over a year. Major tech giants like Nvidia, Salesforce, and Zoom rushed to invest in the company.

At its core, Mercor operates an AI-powered talent marketplace and human data platform that connects top-tier professionals with artificial intelligence labs and global enterprises. Instead of just helping companies hire regular full-time employees, Mercor pays millions of doctors, lawyers, software engineers, and financial analysts to train, test, and grade AI models.

The Simple Explanation of Mercor’s Business Model

To understand how Mercor works, imagine a massive digital bridge connecting two groups of people who desperately need each other.

On one side of the bridge, you have frontier AI companies like OpenAI and major corporations trying to build smart digital workers. These companies have powerful computer chips and mountains of internet text, but their AI models still make silly mistakes when asked to do real professional work. An AI can write a poem in two seconds, yet it might fail miserably at reviewing a complex legal contract or diagnosing a rare medical symptom.

The $0 Secret To Launching Your Entire Online Empire

Why pay hundreds every month for separate software? Build funnels, send emails, and make sales with Systeme.io completely free.

Get Started For $0

Free Forever • No Credit Card Needed • Beginner-Friendly

On the other side of the bridge, you have over five million skilled human experts around the world. These are real doctors, certified accountants, experienced coders, and seasoned marketers looking for flexible, high-paying remote work.

Mercor sits right in the middle of this bridge. Through the official Mercor platform, the company uses its own artificial intelligence to screen, interview, and match these human experts with tech companies. Those experts then complete specialized tasks, write complex prompts, grade AI answers, and build realistic workplace tests so the AI can learn how real jobs are done.

How Mercor Evolved From a Recruiting App to an AI Data Giant

Mercor did not start out as an AI training company. Its journey shows how fast a smart startup can pivot when it spots a bigger opportunity.

Phase 1: Automated Recruiting for Startups

When the founders launched Mercor in 2023, their goal was simple. They wanted to make hiring software engineers faster and cheaper. Back then, startups in the United States struggled to find affordable coding talent, while thousands of brilliant engineers in India and other countries struggled to get noticed by American recruiters.

Mercor built an AI interviewer that could conduct a 20-minute video call with a job candidate, analyze their technical skills, and rank them for employers. If a company liked a candidate, Mercor handled the contract and payroll. It was a clean, efficient recruiting marketplace.

Phase 2: The Pivot to Frontier AI Labs

By late 2024 and early 2025, the founders noticed something surprising. Their biggest customers were not regular software startups looking for full-time coders. Instead, the world’s largest AI labs were using Mercor to hire thousands of temporary specialists.

Why? Because AI models had already read the entire public internet. To make AI smarter, labs needed brand-new, high-level knowledge locked inside the brains of human professionals. Mercor realized its automated vetting system was the fastest way on earth to find, test, and pay thousands of niche experts at once.

Phase 3: Enterprise AI Agents and Simulated Workplaces

By late 2026, Mercor expanded its mission once again. After hitting over $2 billion in gross annualized revenue run rate, the company began selling its human-training infrastructure directly to everyday corporations, not just AI labs.

As reported by The Next Web, Mercor acquired a startup called Deeptune to recreate hundreds of real business software tools as practice playgrounds for AI agents. Today, Mercor helps everyday businesses build custom tests to make sure their internal AI tools actually know how to do the job before they are unleashed on real customers.

The Four Main Ways Mercor Makes Money

Mercor does not rely on a single paycheck. Its business model combines marketplace fees, managed data services, custom testing benchmarks, and enterprise software infrastructure.

1. Hourly Marketplace Markup on Human Experts

This is the engine that drove Mercor’s explosive growth from $1 million to over $2 billion in gross annualized revenue. When an AI lab or enterprise needs 500 licensed physicians or 1,000 senior Python developers for a three-month project, they turn to Mercor.

Mercor charges the corporate client an hourly rate for each expert’s time, and then pays the worker a portion of that fee. Industry estimates show that roughly 60 to 70 percent of every dollar billed goes directly to the human contractor, while Mercor keeps roughly 30 to 40 percent as its net platform fee.

Because Mercor pays out roughly $4 million every single day to its active network of over 30,000 working contractors, that percentage cut turns into hundreds of millions of dollars in net revenue.

2. Reinforcement Learning From Human Feedback (RLHF)

Even if you do not work in tech, you have probably seen RLHF in action. Whenever an AI chatbot gives you two different answers and asks which one you prefer, that is human feedback at work.

For advanced professional tasks, simple thumbs-up or thumbs-down ratings are not enough. AI labs pay Mercor to manage massive RLHF campaigns where domain experts write step-by-step reasoning chains. For example, a corporate lawyer hired through Mercor might spend two hours explaining every single legal risk in a merger agreement so the AI learns how a real attorney thinks.

Send, Spend & Hold Money Worldwide Without Hidden Fees

Transfer money internationally at the real exchange rate and hold 40+ currencies in one account.

Open a Wise Account

Mid-market rate • Fast transfers • 40+ currencies

3. Custom AI Evaluations and Industry Benchmarks (APEX)

How does a company know if an AI model is actually ready to replace or assist a human worker? You have to give the AI a final exam.

Mercor builds high-stakes evaluation tests—called evals—that measure whether an AI can complete multi-step professional projects. Under its APEX benchmark series, Mercor partners with industry leaders to test AI agents in fields like accounting, law, software engineering, and banking. Enterprises pay Mercor to build private versions of these tests tailored to their exact internal rules and software tools.

4. Off-the-Shelf Data Licensing

Aside from custom projects, Mercor also packages verified, high-quality human reasoning datasets and sells access to multiple buyers. When thousands of experts solve complex math, coding, or medical puzzles inside Mercor’s system, those verified solutions become valuable digital assets that AI developers license to improve their models.

Why Mercor’s AI-First Hiring Engine Gives It an Unfair Advantage

You might wonder why a giant company like OpenAI or Google does not just post job ads on LinkedIn and hire these experts directly. The answer comes down to speed, scale, and automation.

Traditional recruiting is painfully slow. Reviewing resumes, scheduling phone screens, conducting interviews, and verifying credentials for just ten people can take a human HR team two months. If an AI lab needs 2,000 specialized chemists by next Tuesday, traditional hiring is impossible.

Mercor solved this bottleneck by automating the entire hiring funnel from start to finish:

  • Automated AI Interviews: Every applicant takes an interactive interview run by Mercor’s proprietary AI. The system asks follow-up questions based on the candidate’s actual answers, testing real depth of knowledge rather than buzzwords on a resume.
  • Instant Skill Grading: Within minutes of finishing the call, Mercor’s software scores the candidate’s communication skills, technical accuracy, and problem-solving speed.
  • Global Compliance and Payroll: Hiring contractors across 100 different countries usually creates a tax and legal nightmare. Mercor acts as the merchant of record, handling contracts, identity checks, and instant payouts in one place.
  • Continuous Quality Tracking: Once an expert starts working on a project, Mercor’s system tracks the quality of their data submissions in real time, automatically flagging careless work and rewarding top performers with higher-paying projects.

Who Are Mercor’s Biggest Competitors?

Operating in a $20 billion market means Mercor does not have the playground all to itself. Several heavyweights fight for the same AI training budgets.

The most famous rival is Scale AI, which reached a $29 billion valuation and pioneered large-scale data labeling. However, after Meta purchased a 49 percent stake in Scale AI for $14.3 billion in 2025, several competing AI labs worried about sharing their secret training plans with a company partially owned by Mark Zuckerberg. Many of those labs shifted massive contracts over to Mercor because Mercor remained independent.

Another major competitor is Surge AI, which also focuses on high-skill human data rather than basic image tagging. Traditional freelance platforms like Upwork and Toptal also overlap with Mercor on the recruiting side, though they lack Mercor’s deep integration into AI model post-training and evaluation environments.

Challenges and Risks Facing Mercor’s Business Model

Rapid growth always brings growing pains, and Mercor’s journey has included several major hurdles that the company must manage carefully.

The $0 Secret To Launching Your Entire Online Empire

Why pay hundreds every month for separate software? Build funnels, send emails, and make sales with Systeme.io completely free.

Get Started For $0

Free Forever • No Credit Card Needed • Beginner-Friendly

Heavy Reliance on a Few Giant Customers

Recent financial reports show that roughly 90 percent of Mercor’s revenue has historically come from a small handful of frontier foundation-model makers. If just two or three giant AI labs decide to cut their spending or build in-house expert networks, Mercor’s revenue could take a sudden hit. This is the exact reason Mercor is spending its new $750 million funding round to win thousands of regular enterprise customers.

Worker Classification and Legal Battles

Like Uber and DoorDash before it, Mercor relies heavily on independent contractors rather than full-time employees. In May 2026, a proposed class-action lawsuit was filed in federal court alleging that Mercor misclassified its expert workers as contractors instead of employees. Managing labor laws across dozens of states and countries remains an expensive ongoing challenge.

Data Security and Rate Cuts

When you manage sensitive training data for the world’s most valuable tech companies, security mistakes are costly. In March 2026, a vulnerability in an open-source software tool exposed some company data, prompting temporary project pauses and legal pushback before Mercor tightened its security controls. Meanwhile, contractors have occasionally pushed back online when project budgets shift or hourly rates get adjusted between contracts.

What Mercor Means for the Future of Remote Work

For everyday professionals around the world, Mercor represents a brand-new way to earn a living. You no longer have to move to San Francisco or London to work on the cutting edge of artificial intelligence.

If you are a nurse in Lagos, an accountant in Manila, a software developer in Bangalore, or a retired professor in Ohio, your real-world expertise has direct monetary value to the companies building the future. Instead of viewing AI purely as a threat that replaces jobs, millions of workers are now getting paid $50, $100, or even $200 an hour to teach AI how their industries actually work.

If you have thoughts on how AI is changing the global job market or want to share a story tip with our editorial team, feel free to reach out through our contact page. You can also join the conversation with our community on Instagram, Facebook, and X.

Frequently Asked Questions About Mercor

What does Mercor actually do?

Mercor is an AI-powered talent marketplace and human data company. It uses artificial intelligence to interview and vet millions of skilled professionals—such as doctors, lawyers, engineers, and writers—and connects them with AI labs and corporations that pay them to train, test, and evaluate AI models.

How does Mercor make money?

Mercor makes money primarily by charging AI labs and enterprise clients an hourly fee for the work completed by its network of experts. Mercor pays roughly 60 to 70 percent of that fee to the contractor and keeps around 30 to 40 percent as its platform margin. It also generates revenue by building custom AI evaluation benchmarks and licensing specialized training datasets.

Who owns Mercor?

Mercor is a privately held company co-founded in 2023 by Brendan Foody, Adarsh Hiremath, and Surya Midha. Following its $750 million Series D round in November 2026, major institutional investors include Benchmark, Felicis, General Catalyst, Nvidia, Salesforce, and Zoom.

Is Mercor a recruiting agency or an AI company?

It is a hybrid of both. While Mercor started as an automated recruiting platform for hiring remote software engineers, its core business today is AI infrastructure—specifically supplying human intelligence, reinforcement learning feedback, and simulated software environments to help AI systems perform real-world jobs.

How much does Mercor pay its experts?

Pay rates on Mercor vary widely depending on the skill level and industry. Generalist data tasks may pay around $16 to $40 per hour, while specialized roles for licensed physicians, senior software engineers, financial analysts, and lawyers frequently pay between $70 and $200 or more per hour.

Final Thoughts

Mercor’s rise to a $20 billion valuation proves one of the biggest ironies of the modern tech era: the smarter artificial intelligence becomes, the more valuable human knowledge gets. By building an automated engine that finds, verifies, and pays millions of real-world experts at lightning speed, Mercor turned human judgment into the essential fuel of the AI economy. As every company on earth races to build its own specialized AI agents over the next few years, Mercor’s position as the middleman between human brains and machine learning looks stronger than ever.

Leave a Comment

Your email address will not be published. Required fields are marked *


Scroll to Top