In the age of digital transformation, integrating data science tools into system design has become pivotal in addressing complex challenges across various fields. Data science tools, such as machine learning algorithms and predictive models, play a crucial role in advancing system design by enabling more efficient and effective solutions. The objective of this paper is to create a technological stack for the implementation of system design using data science tools. The focus is on integrating advanced data analysis and predictive modeling techniques to build an efficient and scalable system. By applying tools for data processing, predictive modeling, and visualization, the paper demonstrates how data science can enhance system design in complex environments. A particular emphasis is placed on using Long Short-Term Memory (LSTM) neural networks for COVID-19 time-series prediction. The developed system is based on a three-layer architecture: a data processing layer, a logic layer with the implementation of LSTM models, and a visualization layer that provides an interactive display of results. The goal is to show how these tools enable accurate forecasts and support decision-makers in understanding complex patterns and predicting future trends. The main contribution of this work is the identification and enhancement of an approach for epidemic prediction, which incorporates best practices and tools for software requirements, design, architecture, verification and validation, testing, maintenance, and evolution. The COVID-19 dataset from Bosnia and Herzegovina is used as a case study. The results are expected to demonstrate how artificial intelligence, data analytics, and big data, when applied in the context of software engineering, can support data analysis and predictive modeling to facilitate informed decision-making.