Data Science & Analysis - AI Training - Sheffield, UK

Role overview Apply data science or analysis expertise to review and improve AI-generated technical work. Assignments include assessing statistical reasoning, data insights, code, and predictive modelling, with attention to correctness, clarity, and adherence to task guidelines. Candidates complete a short skills assessment before being invited to participate in paid tasks.

Responsibilities - Evaluate AI-generated data insights and statistical analyses for accuracy, clarity, and completeness. - Check mathematical and statistical claims against authoritative references. - Run data-related Python or R scripts to verify results and test edge cases. - Assess how models reason through data problems and build predictive models, documenting strengths and weaknesses. - Review responses for compliance with specified guidelines and professional standards.

Requirements - A bachelor’s, master’s, or doctoral degree in data science, statistics, mathematics, or a related quantitative field. - Practical experience in data analysis, predictive modelling, or machine-learning engineering. - Ability to solve complex statistical problems and independently interpret advanced data visualisations. - Strong attention to detail, including the ability to spot flawed reasoning in statistical arguments and data interpretations. - Substantial experience using large language models in data workflows, with an understanding of their common failure modes. - Clear communication of mathematical concepts and analytical findings; a PayPal account is needed to receive task payments.

Nice to have - Proficiency in Python, including common data and machine-learning libraries, or R; advanced SQL; and data visualisation tools such as Tableau, Power BI, Matplotlib, or Seaborn. - Knowledge of machine-learning frameworks such as PyTorch or TensorFlow and statistical modelling methods.

Benefits and work setup Tasks are paid, with advertised rates of up to $80 per hour; rates may vary by assignment. Some work requires one uninterrupted hour, though many tasks are shorter. The arrangement is described as flexible and remote, subject to available assignments.