Developed a multilingual OCR system using Python and PyTorch that performs document layout analysis and extracts text with high accuracy. Conducted experiments and visualizations in a Google Colab notebook, demonstrating end-to-end OCR pipeline capabilities and performance evaluation.