Large Language Models (LLMs), have taken the world by storm, demonstrating their ability not only to automate tedious tasks, but also to show some degree of proficiency in completing software engineering tasks. A key concern with LLMs is their "black-box" nature, which obscures their internal workings and could lead to societal biases in their outputs. In the software engineering context, in this short paper, we empirically explore how well LLMs can automate recruitment tasks for a geographically diverse software team. We utlize OpenAI’s GPT to conduct an initial set of experiments using GitHub Profiles from four regions (i.e., the United States, India, Nigeria, and Poland) to recruit a six-person software development team, analyzing a total of 3,896 profiles over a 5-year period (2019–2023). The results indicate that GPT tends to prefer some regions over others, even when some profiles have been manipulated to contain counterfactuals, such as swapping the location strings of two profiles. Furthermore, GPT was more likely to assign certain developer roles to users from a specific country, revealing an implicit bias. Overall, this study reveals insights into the inner workings of GPT and has implications for mitigating these potential biases.

This repository serves as the online appendix for the paper "Nigerian Software Engineer or American Data Scientist? GitHub Profile Recruitment Bias in Large Language Models".