Remote MPH: Health Policy & AI Data Trainer

Alignerr Corp.; Manchester; Remote; GBP 15,000 - 30,000; Part time

Job summary

Alignerr Corp. is seeking a Health Policy & Management - AI Data Trainer to help ensure health information used by AI is accurate and evidence-based.

The role is fully remote and offered as an hourly contract, with flexible hours to fit 10-40 hours per week. You will design public health scenarios, author ground-truth solutions from peer-reviewed research and guidelines, and audit model outputs for bias and accuracy.

Qualifications

• Holds MPH or MSPH or a closely related graduate degree.
• Strong foundational knowledge in at least two of: Epidemiology, Biostatistics, Environmental Health, or Health Policy.
• Able to translate complex health data into clear, actionable content.
• Precise and detail-oriented when evaluating significance, data citations, and logic.

Responsibilities

• Design Complex Public Health Scenarios - Build advanced case studies involving disease outbreak simulations, health equity assessments, and statistical interpretation of clinical trial data.
• Author Ground-Truth Solutions - Write rigorous, step-by-step responses synthesizing peer-reviewed research, epidemiological data, and guidelines (CDC/WHO).
• Audit AI Outputs - Evaluate AI-generated health advice for accuracy, bias, and ethics; identify misinformation and gaps in reasoning.
• Refine AI Reasoning - Provide feedback to distinguish correlation from causation and improve health risk communication.

Skills

MPH or MSPH

Epidemiology

Biostatistics

Health Policy

Education

MPH or MSPH

Tools

R

SAS

Stata

Job description

Alignerr Corp. is seeking a Health Policy & Management - AI Data Trainer to help ensure health information used by AI is accurate and evidence-based.

The role is fully remote and offered as an hourly contract, with flexible hours to fit 10-40 hours per week. You will design public health scenarios, author ground-truth solutions from peer-reviewed research and guidelines, and audit model outputs for bias and accuracy.