You will review U.S. tax filing workflows and related financial processes used in AI training. You will compare system-generated outputs with tax documents and realistic filing scenarios, then assess accuracy, completeness, and ease of use.
The work includes document annotation, rubric-based evaluation, and clear feedback. Collaboration is asynchronous, with timely updates on assigned deliverables.
- Review and annotate U.S. tax filing workflows and related financial processes.
- Compare system-generated outputs with tax documents and realistic filing scenarios.
- Apply defined evaluation criteria consistently and objectively.
- Provide clear written and verbal feedback on your findings.
- Use relevant financial accounts, Gmail, and Google Drive when required for workflow reviews and document sharing.
- Review tax documents only after personal information has been sanitized.
What It Pays and Takes
This is a remote contractor project with part-time work and asynchronous collaboration. Prior AI training experience is not required, although experience with data annotation, quality review, rubric-based assessment, or model evaluation is helpful.
- Pay: $30 to $110 USD per hour.
- Time requirement: 20 or more hours per week.
- Location: You must live in a mainland U.S. state.
- Language: English.
- Experience: Hands-on experience filing 2025 U.S. taxes independently or with a CPA or other tax professional.
- Skills: Financial literacy, strong attention to detail, and consistent document review and annotation.
- Evaluation: Ability to apply defined rubrics objectively.
- Communication: Clear written and verbal communication skills.
- Tools: Familiarity with Google Docs, Sheets, and Drive.
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 learn from examples, including reviewing documents, rating outputs, and checking whether responses follow real-world rules. People with financial and tax expertise help make these systems more accurate and useful.