Similar Jobs
See allData Science & Analysis - AI Training
Prolific
Global
Python
R
SQL
Software Engineer, Benchmarking
Office Hours
US
Python
Docker
React
Research Intern
Cohere
Global
Python
Machine Learning
NLP
Head of AI Research
A1
Global
Python
PyTorch
JAX
Member of Technical Staff
AI Digest
Global
Data Analysis
Software Engineering
AI Safety
About Mercor:
- Mercor organizes human intelligence to power the AI economy by building between human expertise and frontier models.
- It is a leading AI data company with millions of domain experts paid over $4 million per day to train AI models.
- Mercor is a profitable Series C company valued at $10 billion with in-person offices in San Francisco, NYC, and London.
About the Fellowship:
- The Mercor Research Fellowship funds people to build the next generation of benchmarks and evaluation techniques.
- You pitch a benchmark or eval methodology and, if selected, receive time, compute, expert labor, and mentorship to design and release it.
- The fellowship is 3–6 months, remote or in-person in San Francisco, with a stipend of $40,000 or $80,000.
What You’ll Do:
- Propose and scope a new benchmark or evaluation technique in a domain APEX doesn't yet cover.
- Design, build, and validate the benchmark with domain experts, including task specifications and scoring.
- Run frontier models against your benchmark, analyze failures, and publish results as a paper or dataset.
Compensation Benefits:
- Unlimited API credits and dedicated GPU compute budget.
- Weekly 1:1 mentorship with APEX research team.
- Optional desk in Mercor’s San Francisco office and network introductions.
Mercor
Mercor organizes human intelligence to power the AI economy by building the layer between human expertise and frontier models. It is a profitable Series C company valued at $10 billion with a culture of in-person collaboration in San Francisco, NYC, or London.