Let's Get You Hired: A Job Seeker's Perspective on Multi-Agent Recruitment Systems for Explaining Hiring Decisions

Transparency in hiring decisions is valued by job seekers: The study found that participants perceived the multi-agent LLM system as "significantly more trustworthy and fair" compared to traditional recruitment methods, suggesting that candidates want clear explanations for why they are or aren't selected for positions.

Explainability makes AI systems more actionable: Beyond just transparency, job seekers found the system's explanations "significantly more actionable," indicating that candidates want not only to understand hiring decisions but also to receive guidance on how they can improve their applications and candidacy.

User-centered design is critical for responsible AI in hiring: The researchers used iterative design with actual job seekers (first with 4 participants, then 20) to develop the system, demonstrating that involving affected users throughout the development process—rather than designing in isolation—leads to AI systems that better serve people's needs and concerns.

Multi-agent AI systems can address limitations of black-box hiring tools: The paper highlights that traditional Applicant Tracking Systems (ATS) lack transparency, and shows that a more sophisticated AI approach using multiple agents can provide job seekers with the explanations and fairness they currently lack in automated recruitment.

Design principles from this work extend beyond hiring: The authors note their findings offer "broader design implications for building user-aligned, multi-agent explainable AI systems across diverse domains," suggesting these lessons about transparency, actionability, and fairness could improve responsible AI implementation in other high-stakes decision-making contexts.