Description

You will use your statistics expertise to train large language models by challenging them on topics like statistical inference, model selection, and causal inference. You'll document error modes to improve model reasoning and help shape the next generation of AI.

Requirements

• Master's or PhD in statistics, data science, mathematics, or related field
• Peer-reviewed publications or applied statistical research preferred
• Hands-on experience with large datasets
• Clear, metacognitive communication skills

Responsibilities

• Challenge language models on statistical inference, model selection, sampling methods, causal inference, ANOVA, and statistical power
• Verify factual accuracy and logical soundness of model outputs
• Capture reproducible error traces from model failures
• Suggest improvements to prompt engineering and evaluation metrics