A Research Study on Agentic AI and AI Agents: Transforming Organizational Automation and Decision-Making

This study examines how Agentic AI and multi-agent architectures are reshaping organizational automation, recruitment intelligence, and workforce decision-making. Drawing on a multi-agent Retrieval-Augmented Generation (RAG) system integrated with Large Language Models (LLMs), the research evaluates autonomous AI agent pipelines across two enterprise HR functions: recruitment shortlisting and employee attrition prediction. Three publicly available datasets totalling 5,154 records were used to train, evaluate, and compare the proposed system against conventional machine learning baselines. The multi-agent RAG-LLM architecture achieved an F1 score of 0.89 and AUC-ROC of 0.94 for candidate shortlisting, and 91.2 percent accuracy with an F1-score of 0.83 for attrition prediction, substantially outperforming keyword-based and single-model approaches. The findings affirm that Agentic AI, when governed through explainability mechanisms and human-in-the-loop safeguards, produces measurably superior and ethically defensible organizational outcomes compared to legacy automation methods.