Automated document classification and information extraction are important tasks in many industries, including healthcare, finance, and government. In healthcare, patient data is stored in unstructured formats, which makes data retrieval and analysis difficult. In the finance sector, financial documents such as data analysis, regulatory compliance, and fraud detection are also important tasks. In government, the classification and extraction of information from legal and policy documents can enhance decision-making and policymaking processes. Automated document classification and information extraction using computer vision and deep learning, can efficiently extract important information from these unstructured documents, such as medical histories, diagnoses, treatments, insurance applications and improve the process of data analysis, fraud detection, and decision-making process. By using this, businesses and organizations can reduce the time and cost associated with manual document processing, enabling them to focus on more critical tasks. This paper proposes a solution that utilizes OCR, NLP and CNNs to perform these tasks on insurance applications. The system is trained on a large dataset of documents and can classify and extract information from various types of documents, including invoices, receipts, and forms. Our experiments show that the proposed system achieves high accuracy of 90% in document classification and information extraction tasks, making it suitable for use in realworld applications.