Nicolai Olsen contributed backend improvements to the CogitoNTNU/DeepTactics-Muzero repository, focusing on enhancing code observability and correctness within the training loop. He implemented detailed logging for training losses, enabling tracking of total policy, reward, and value losses, and introduced optional aggregated loss outputs to support debugging and metric interpretation. Using Python and leveraging his skills in deep learning and debugging, Nicolai also addressed a duplicate calculation in the loss function, adding clarifying comments to improve maintainability and prevent future logging errors. His work strengthened the reliability and traceability of the training process, reflecting a thoughtful and methodical engineering approach.