This study applies a Long Short-Term Memory (LSTM) neural network to forecast stock closing prices for selected technology companies (Apple, Google, Microsoft, and Amazon). The paper documents data collection, preprocessing, exploratory analysis (returns, volume, correlations), model architecture, and results. The aim is to evaluate LSTM’s ability to capture temporal patterns in stock prices and to provide practical insights for short-term forecasting. Key findings show that the LSTM model captures overall price trends and produces reasonable short-horizon forecasts; however, prediction accuracy is affected by market volatility, data noise, and model complexity.