About Mercor

Mercor’s mission is to organize human intelligence for the AI economy. We build the infrastructure that enables frontier AI labs and enterprises to train, evaluate, and improve AI on real-world professional work. More than five million experts have joined our network, and we pay out over $5 million daily for their knowledge and expertise.

We built our training and evaluation infrastructure alongside the world’s leading AI labs. Now, we’re bringing those capabilities to enterprises across industries, enabling them to train, evaluate, and improve AI using their own expertise and standards. Our APEX benchmarks measure how AI performs on economically valuable work, while our enterprise evaluations and post-training capabilities turn those insights into measurable improvements across real-world workflows.

Our work sits at the intersection of frontier AI, human expertise, and real-world application. We’re solving problems few companies have encountered, with the opportunity to influence what the world’s most advanced AI systems become capable of doing. We’re looking for people who bring different perspectives, take ownership, exercise sound judgment, and hold themselves to an exceptionally high standard. You’ll have the opportunity to do consequential work alongside talented colleagues who challenge one another, collaborate openly, and share a commitment to excellence.

Mercor is profitable and recently raised a $750 million Series D at a $20 billion valuation. We work in person five days a week in San Francisco, New York, and London.

About the Team

Frontier models have been trained on most of the public web. The next gains will come from how work actually gets done inside companies: the documents, spreadsheets, code, email, chat, and business systems where real decisions get made, each with its own structure and quirks. That data is private, messy, and inconsistent, and turning it into training data that labs can use and trust is an unsolved problem. The Data Acquisition team works on that problem, turning real-world data into datasets that frontier labs use to train and evaluate their models. Our work spans domains from healthcare and engineering to finance and legal. The same data is valuable at every stage of training, from pre-training to the environments where agents learn to do real work.

About the Role

As a Product Manager for Real World Data, you will own how real-world data becomes datasets labs want to buy. You will define what makes this data useful at each stage of training, decide how datasets are shaped, packaged, and delivered, and get high-quality data into labs' hands faster. The role spans product and go-to-market: you will need to understand what is in the data, what labs need, and how to sell it. Working this close to real work, you will see firsthand where frontier models fall short on the tasks that matter most. Real-world data is a new category of product, and what you build will help define it and shape what the next generation of models learns from.

In this role, you will

Build New Product Lines: Partner with research and environments teams to make real-world data useful for post-training, so agents learn the messy, judgment-heavy work they will face in real workplaces.

Own the Roadmap: Define success in terms of sellable datasets, and use lab feedback, sales outcomes, and unmet demand to decide what the team builds next.

Lead Product Discovery: Define what makes real-world data valuable for pre-training, mid-training, and post-training, and turn that understanding into dataset specifications, quality standards, and ways to measure impact on models.

Define Packaging and Positioning: Turn acquired data into off-the-shelf dataset products, deciding what is sellable, to whom, and in what form.

Set the Quality Bar: Set standards for how datasets are structured, documented, and delivered, and build the visibility to catch problems before delivery and trace each one to its source.

Improve Time to Value: Shorten the path from customer request to accepted delivery by tightening scoping, sampling, and quality review.