Research Information System
5th International Conference on Informatics and Software Engineering, IISEC 2026, Ankara, Turkey, 5 - 06 February 2026, pp.645-650, (Full Text)
Retirement planning is a complex decisionmaking process that requires the integration of personal, financial, and health-related factors. This study introduces RetireSense AI, a modular Python-based application enhanced by large language models (LLMs) and function-calling architecture. The tool estimates optimal retirement age, predicts healthcare costs, and analyzes longevity risk using dedicated AI agents. Each agent is triggered through structured function calls rather than probabilistic generation, ensuring deterministic and explainable results. The system converses naturally with users to gather necessary input and performs computations using predefined algorithms. Evaluation shows that ninety-five percent of queries are processed within three seconds, with ninety-nine percent of data consistency and high usability confirmed by post-session feedback. The application demonstrates potential as a reliable, data-driven decision support system for personalized retirement forecasting.