Chaitanya Teegala

From Human-Driven Tuning to Agentic AI: Automating Hyperparameter Optimization in XGBoost

This talk compares manual XGBoost hyperparameter tuning with an agent-based approach that systematically explores parameter space, evaluates performance, and tracks historical experiments for transparent analysis. The session covers architecture, evaluation strategy, and outcome comparisons to show how agentic AI improves reproducibility and accelerates model optimization.