NARUTO is a neural active reconstruction system that combines a hybrid neural representation with uncertainty learning to enable high-fidelity surface reconstruction. The system leverages a multi-resolution hashgrid as the mapping backbone, chosen for its fast convergence and ability to capture high-frequency local features. The core of NARUTO is an uncertainty learning module that dynamically quantifies reconstruction uncertainty while actively reconstructing the environment. This module is used to propose a novel uncertainty aggregation strategy for goal searching and efficient path planning. NARUTO enables 6DoF movement in unrestricted spaces, enhancing the completeness and quality of scene reconstruction. Extensive evaluations using an indoor scene simulator and benchmark datasets like Replica and MP3D demonstrate superior performance, setting a new standard in active reconstruction. The system's effectiveness is further validated through ablation studies and comparisons with existing methods, highlighting its robustness and practicality. Future work aims to address real-world challenges by improving localization, expanding motion constraints, and enhancing uncertainty representation.NARUTO is a neural active reconstruction system that combines a hybrid neural representation with uncertainty learning to enable high-fidelity surface reconstruction. The system leverages a multi-resolution hashgrid as the mapping backbone, chosen for its fast convergence and ability to capture high-frequency local features. The core of NARUTO is an uncertainty learning module that dynamically quantifies reconstruction uncertainty while actively reconstructing the environment. This module is used to propose a novel uncertainty aggregation strategy for goal searching and efficient path planning. NARUTO enables 6DoF movement in unrestricted spaces, enhancing the completeness and quality of scene reconstruction. Extensive evaluations using an indoor scene simulator and benchmark datasets like Replica and MP3D demonstrate superior performance, setting a new standard in active reconstruction. The system's effectiveness is further validated through ablation studies and comparisons with existing methods, highlighting its robustness and practicality. Future work aims to address real-world challenges by improving localization, expanding motion constraints, and enhancing uncertainty representation.