Mercor 研究员计划 — APEX
Mercor Research Fellowship — APEX
Mercor的使命是组织人类智能，为人工智能经济提供动力。我们是一家领先的AI数据公司，构建了连接人类专业知识和前沿模型的层级。平台上数百万领域的专家每天获得超过400万美元的报酬，用于训练前沿AI模型。Mercor的APEX基准系列衡量AI在现实世界中的实际影响，包括投资银行和公司法中的多小时代理任务、真实的职场会计工作流程、真实世界的软件工程以及研究生水平的科学。每个APEX基准都是通过Mercor的领域专家网络构建和验证的，而不是从教科书中编写出来的。
Mercor的企业部门将这种基础设施带给财富500强公司：帮助公司捕捉其优秀员工的实际工作方式，并将这些专业知识直接转化为代理系统。
Mercor正在创造一种新的工作类别，其中专业知识推动AI的进步。实现这一点需要一个雄心勃勃、快节奏且 deeply committed 的团队。你将与研究人员、运营人员和处于塑造重新定义社会的系统的最前沿的AI公司合作。Mercor是一家盈利的C轮融资公司，估值达100亿美元。我们在旧金山、纽约或伦敦的办公室每周五天亲自办公。
关于研究实习计划
Mercor的APEX基准系列衡量前沿AI模型是否真的能完成具有经济价值的工作：投资银行和公司法中的多小时代理任务、真实的职场会计工作流程、真实世界的软件工程以及研究生水平的科学。每个APEX基准都是通过Mercor的领域专家网络构建和验证的，而不是从教科书中编写出来的。
除了APEX，Mercor的经济学团队致力于理解AI如何改变劳动力市场，并尝试量化其经济影响。为此，我们利用Mercor独有的数据访问权限，包括：拥有数百万候选人的市场、内部AI使用指标、企业数据等。
Mercor研究实习计划资助人们开发下一代基准测试、评估技术和AI的经济分析。你可以提出你想构建的基准或评估方法——一个新的领域、一个更复杂的任务格式、一种更好的测量代理可靠性的方式——如果被选中，你将获得时间、计算资源、专家劳动力和指导，以从头开始设计、实施并发布它。或者，在经济学方面，你可以提出一项使用Mercor专有数据的研究，以更好地理解AI对经济和企业的影响的某个方面。
你将直接与APEX研究团队合作，获得来自Mercor的财富500强和前沿实验室合作伙伴的真实企业评估问题，并看到你的工作如何塑造行业的发展。
查看英文原文
ABOUT MERCOR
Mercor's mission is to organize human intelligence to power the AI economy. We're a leading AI data company, building the layer between human expertise and frontier models. Millions of domain experts on the platform are paid over $4 million per day to train frontier AI models. Mercor's APEX benchmark family measures AI's real-world impact on professional work. Mercor Enterprise brings this same infrastructure to Fortune 500 companies: helping companies capture how their best people actually work, translating that expertise directly back into agents.
Mercor is creating a new category of work where expertise powers AI advancement. Achieving this requires an ambitious, fast-paced and deeply committed team. You’ll work alongside researchers, operators, and AI companies at the forefront of shaping the systems that are redefining society. Mercor is a profitable Series C company valued at $10 billion. We work in-person five days a week in our San Francisco, NYC, or London offices.
About the Fellowship
Mercor’s APEX benchmark family measures whether frontier AI models can actually do economically valuable work: multi-hour agentic tasks in investment banking and corporate law, real professional accounting workflows, real-world software engineering, and graduate-level science. Every APEX benchmark is built and validated with Mercor’s network of domain experts — not written from a textbook.
Alongside APEX, Mercor's economics team works to understand how AI is changing the labor market and tries to quantify its economic impacts. To do so, we leverage Mercor’s unique access to data, including: a marketplace with millions of candidates, internal AI usage metrics, enterprise data, and more.
The Mercor Research Fellowship funds people to build the next generation of benchmarks, evaluation techniques, and economic analyses of AI. You can pitch a benchmark or eval methodology you want to build — a new domain, a harder task format, a better way to measure agentic reliability — and if selected, you get the time, compute, expert labor, and mentorship to design, implement, and release it end to end. Or, on the economics side, you can pitch a study using Mercor’s proprietary data to better understand an aspect of AI’s impact on the economy and firms.
You’ll work directly with the APEX research team, get access to real enterprise evaluation problems from Mercor’s Fortune 500 and frontier-lab partners, and see your work shape how the industry measures AI capability.
Program Details
- Duration: 3–6 months, rolling admission
- Commitment: minimum 30 hours/week; full-time preferred
- Location: remote, or in-person at Mercor’s San Francisco office
- Admission: apply with a specific benchmark or eval technique you want to build — the fellowship is funded around your pitch, not a generic research rotation
What You’ll Do
- Propose and scope a new benchmark evaluation technique, or economic study in an area not covered well by existing work
- Build and validate it. For benchmarks, that means designing task specifications and grading rubrics with Mercor's network of vetted experts — lawyers, accountants, engineers, scientists, consultants — then piloting tasks, calibrating scoring, and running frontier models to analyze where and why they fail. For economic studies, this can mean anything from designing and running field experiments with experts on our platform to conducting empirical analyses of enterprise data.
- Publish your results: a paper, an open dataset, a new leaderboard on APEX, or a methodology the team adopts internally — and partner with Mercor's research and engineering teams to fold what you learn back into our public work.
Focus Areas
- Long-horizon, multi-app agentic tasks in professional services (law, finance, consulting) — extending APEX-Agents
- Real-world software engineering evaluation beyond issue resolution — extending APEX-SWE
- Professional accounting and finance workflows — extending APEX-Accounting
- AI-for-Science evals: research-level mathematics, biology, materials science, and theoretical physics
- New economic benchmarks — negotiation, management, and other capabilities that carry economic value but resist standard task formats
- Novel evaluation methodology: contamination resistance, rubric design, human-vs-model grading agreement, cost-adjusted scoring
- The economics of human data platforms like Mercor – pricing, matching, elasticity of supply and demand.
- AI Theory of the Firm: what an AI-native firm looks like from the inside, and how work is actually organized within one, with a focus on empirical work.
- Strong pitches outside this list are welcome — we fund the best ideas, not the closest fit to a template.
What We’re Looking For
- Genuine interest in evaluation as a research discipline — not just a stepping stone to a model-building role.
- Background in economics, CS, ML, statistics, or an adjacent field (measurement, psychometrics, HCI, social science)
- A specific, well-scoped idea for a benchmark, eval technique, or empirical study you want to build — the fellowship is built around your pitch.
- Comfortable in a startup environment: fast iteration, direct access to real customer problems, less hand-holding than an academic lab.
- Able to commit at least 30 hours/week for the duration of the fellowship — during a leave, over a summer, or a flexible stretch of a PhD.
- Bonus: experience with agentic evaluation, RL environments, or domain expertise in law, finance, medicine, or a scientific field.
Compensation & Benefits
- 3-month stipend of $40,000 or 6-month stipend of $80,000
- Unlimited API credits, plus a dedicated budget for GPU compute and paid expert/human-data time
- Weekly 1:1 mentorship with a member of the APEX research team, plus regular access to the broader research org
- Access to Mercor's marketplace and expert-network data for empirical work, subject to review
- Access to frontier model APIs, Mercor’s internal evaluation infrastructure, and — where appropriate — real enterprise evaluation problems from Mercor’s customers
- Optional desk in Mercor’s San Francisco office for fellows who want to be in person
- Introductions to Mercor’s network of researchers across frontier labs and academia
- Standout fellows are considered for a full-time offer on the APEX research team at the end of the fellowship