Recent developments in hadronic object reconstruction and calibration in ATLAS have led to significant improvements in jet performance. Advances at the jet input level, including enhanced topo-cluster performance through cell timing cuts and novel machine learning techniques, enable reconstructing jets with improved precision. Innovative approaches, including the use of single-particle measurements to calibrate the jet energy scale, the use of pile-up collisions to improve the precision of the jet energy resolution at low transverse momentum, and machine-learning–based techniques to correct jets originating from bottom quarks, provide unprecedented accuracy in jet measurements. These advancements promise substantial benefits for a wide range of forthcoming analyses involving jets.