The work
You will help improve AI systems by creating difficult quantitative problems and reviewing the answers they produce. Your feedback should show whether an answer is technically accurate, practical, and consistent with the methods used by working quants.
The work draws on research, modeling, strategy development, and technical reasoning across quant research, signal generation, trading, risk, and portfolio settings.
- Create expert-level quantitative problems and scenarios based on research, modeling, and strategy work.
- Review AI-generated quantitative responses for accuracy and practical feasibility.
- Assess whether responses meet the standards used by practicing quantitative analysts.
- Write structured feedback to help improve AI model performance.
- Apply judgment across signal generation, trading strategy design and execution, risk modeling, and portfolio construction.
- Complete project tasks independently in a remote, asynchronous setting.
What it pays and takes
This is a flexible, project-based contractor role for candidates who can work independently and explain complex quantitative ideas clearly.
- Pay: Up to $150 per hour.
- Time: 20+ hours per week.
- Location: Open to candidates based in the United States.
- Work setting: Remote and flexible.
- Professional background: At least one year of hands-on experience working as a quant at a top company or fund.
- Education: A bachelor's degree in mathematics, statistics, computer science, or a related quantitative field.
- Expertise: Depth in quant research and signal generation, trading strategy design and execution, or risk modeling and portfolio construction.
- Skills: Strong written communication, close attention to detail, and the ability to explain complex quantitative concepts clearly.
- Helpful background: Experience at a recognized, top-tier quantitative fund or trading firm.
How it works
Apply on OpenTrain with your resume, then complete the application on the hiring site.
About AI training work
AI training is the human work behind systems that generate and evaluate text, code, images, and other outputs. People with subject expertise are needed to create realistic examples, judge technical answers, and explain what a better response should contain.