Skip to content

By playing, you agree to uListen's Terms and Privacy Policy and YouTube's Terms of Service.

Sourcery with Molly O'SheaSourcery with Molly O'Shea

Mercor Raises $750M at $20B Valuation | CEO Brendan Foody

Mercor has raised $750 million in a Series D at a $20 billion valuation, with participation from Nvidia, Benchmark, General Catalyst, Funds managed by affiliates of Apollo, XTX Ventures, Salesforce, and others.

Brendan FoodyguestMolly O'Sheahost
Nov 10, 202654mWatch on YouTube

Summaries, chapters, transcripts, and FAQs on this page are made by uListen, not YouTube.

Great conversations run long. Their best ideas don't have to.

uListen is free on iPhone, iPad, Android, Chrome, and Claude.

See how

At a glance

WHAT IT’S REALLY ABOUT

Mercor’s playbook: scaling experts, evals, and RL environments for agents

  1. Brendan Foody describes Mercor’s mission as building the systems for how humans contribute economically by teaching and evaluating agents rather than performing repetitive tasks.
  2. He outlines Mercor’s three-product stack—talent platform, Studio data platform, and AI project manager—designed to create reinforcement learning environments for labs and AI companies.
  3. He explains why Mercor should not be viewed as a marketplace, emphasizing that the deliverable is an RL environment made of apps, worlds, and tasks, with experts as one component of the overall system.
  4. He argues the category is durable because models need humans to define and measure the frontier via rubrics and feedback, and making humans more productive increases demand for teaching models.
  5. He predicts enterprises will increasingly build internal eval and post-training capabilities as inference spending grows, and he frames data sovereignty as critical to preventing frontier labs from subsuming competitive advantage across industries.

IDEAS WORTH REMEMBERING

5 ideas

Mercor is oriented around how humans teach agents, not how humans do repetitive work.

He frames Mercor as building the tooling and workflows for a world where humans contribute less by doing repetitive tasks and more by training agents, creating evals, updating “skills files,” and building reinforcement learning environments that mirror real work.

Mercor’s stack is three products that together produce RL environments at scale.

He describes three core products: a talent platform (over five million people looking for work), “Studio” as a data platform with an AI co-pilot and agentic quality control for building RL environments, and an AI project manager that automates managing experts and data quality to hit customer outcomes.

Operational reliability—especially payments—was treated as a core moat after early failures elsewhere.

He says a major catalyst for building world-class payments and direct relationships with labs came from seeing experts not get paid on time when Mercor supplied talent through another player, leading Mercor to prioritize being accountable for expert experience and payouts.

The product Mercor sells is RL environments, with experts as one ingredient—not the whole business.

He argues Mercor isn’t a marketplace taking a simple matching fee; the actual deliverable is an RL environment composed of apps (cloned tools), worlds (the content inside), and tasks (prompts and rubrics that produce grades/rewards for learning).

Greater human productivity in data creation increases total demand for human teaching, not less.

He says agents and AI copilots increase expert productivity (he estimates 3–5x), but demand expands rather than shrinks (he references Jevons paradox), because most knowledge is still in people’s heads and training agents becomes the scalable alternative to repetitive labor.

WORDS WORTH SAVING

5 quotes

Darsh and I have had opportunities where we could walk away with a lot of cash, but that's not what motivates us. Like, we'd rather work 100 hours a week to solve a really important problem in the world.

— Brendan Foody

The barrier to automating the entire economy is, how do we cover the full distribution of all of the RL environments that mirror everything in the distribution of the economy?

— Brendan Foody

Ultimately, like, 99% of, of knowledge still lives in people's heads, and if we just make it more effective and productive to have humans teaching models, that just increases the amount that people teach models throughout the economy.

— Brendan Foody

I think founders gain an appreciation that, like, chaos is the norm.

— Brendan Foody

I think that the key development is that every Fortune 500 is going to look like a Neo lab.

— Brendan Foody

Read the available transcript
Shift from repetitive work to training agentsMercor’s three-product structureStripe Connect payouts and payments reliabilityRL environments: apps, worlds, tasksAgents, copilots, and Jevons paradox demand expansionExpert sourcing via referrals and AI interviewingCategory durability and “frontier” measurement by humans (evals/rubrics)

Ask ChatGPT about this episode

Get its key ideas, plus a summary of each chapter linked to that point in the video, right in your chat.

  1. Add uListen AI to ChatGPT.

  2. Paste this prompt into a new chat.

    Use uListen AI to give me this episode's key ideas and a chapter-by-chapter summary, with timestamps: https://www.youtube.com/watch?v=xzA6lYBzmzQ

SPEAKERS

  • Brendan Foody

    guest

    CEO of Mercor, an AI-driven talent and evaluation platform.

  • Molly O'Shea

    host

    Host of the interview podcast Sourcery.

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

Read along on your phone, search it beside the video, or ask your favorite AI app what was said.