The company is seeking statistics experts to train large-scale language models by engaging in technical dialogues and verifying the accuracy of statistical reasoning. Candidates should ideally hold a Master's or PhD in a quantitative field and be proficient in areas like probability theory, Bayesian inference, and machine learning. The role involves identifying model failures and suggesting improvements to help harden AI reasoning in scientific contexts.