Which piece do you start with in a puzzle without an end? The aim of this study was to investigate whether the analysis of cybercrime can be streamlined using moderntechnologies and tools. A quantitative experimental method was employed, analyzing an existing dataset. Through both manual analysis and language model-based processing, the study examined which digital traces are most useful for reconstructing cyberattacks. The results showed that certain artifacts could be linked to specific attacks. The language models demonstrated varying capabilities in analysis and filtering tasks, ranging from high accuracy to an inability to perform the analysis. The conclusion drawn is that timeline reconstruction can be made more efficient through careful selection of artifacts and adaptation to the nature of the attack. AI models can support the analytical process, but they still require human oversight and must be aligned with forensic standards. The study highlights both the potential and limitation of current technologies and emphasizes the need for continued research in this field. Particularly important is the development of robust frameworks to ensure data integrity and legal defensibility in automated analyses. The hope is that future innovations will reduce the workload for IT forensics professionals while improving the reliability of cybercrime investigations.