AI Trainer, Data Science

Brunel • New York, NY

1d • $40 - $125 • Remote

About The Position

Brunel is partnering with DataAnnotation to connect experienced data scientists with an AI training program. Experts use hands-on data science knowledge to evaluate and improve how AI models handle real data science work, contributing to high-quality, technically accurate AI systems. The work spans product and business analytics, applied machine learning, statistics and experimentation, forecasting, analytics engineering, and applied NLP or computer vision.

This is fully remote, project-based work with a flexible schedule. The minimum commitment is 10 hours per week, with no weekly maximum; full-time hours are available. The pay rate is $40–$125 USD per hour, paid via PayPal. Applicants must be located in Australia, Canada, Ireland, New Zealand, the United Kingdom, or the United States. The first step is a skills assessment lasting 2–4 hours. This is an independent contractor position.

Requirements

• Native- or bilingual-level fluency in English.
• At least 2 years of professional applied data science experience in industry or research.
• Working proficiency in Python and/or SQL, including the ability to write, debug, and explain analysis code.
• Experience owning a multi-step analysis end to end, from raw data through modeling and interpretation to a decision.
• Ability to judge whether a result is valid and explain why an answer is wrong, not merely identify that it is wrong.
• Comfort working with AI tools, or ability to learn them quickly.

Nice To Haves

• A current, in-progress, or completed Bachelor's degree in a quantitative field is preferred.

Responsibilities

• Design challenging, realistic data science tasks based on professional practice, such as exploratory analysis of messy data, A/B test design and readouts, or properly validated models.
• Run tasks through frontier AI models and assess the outputs against professional standards.
• Compare outputs from two models on identical prompts and files, decide which performed better, and document each output's shortcomings.
• Write detailed grading rubrics and tests that define correct deliverables and explain why a response passes or fails.
• Identify specific failures with evidence, including leakage, confounding, incorrect joins, misapplied tests, code that does not fulfill the request, fabricated or ignored data, and off-brief interpretation.
• Contribute across analytics, machine learning, experimentation, and data engineering.
• Review and refine tasks developed by other experts.

What This Job Offers

• Career Level: Mid Level
• Education Level: No Education Listed