Data Science Intern (Document Classification and Anomaly Detection)
Built OCR-based document processing pipelines using Azure Document Intelligence and LLM post-processing to automate high-accuracy document classification. Designed end-to-end workflows for structured data extraction from large volumes of documents and implemented automated discrepancy detection using AI. Utilized hybrid methods to enhance anomaly identification and reduce manual workload significantly. • Created automated pipelines for document ingestion and processing. • Leveraged OCR and LLMs for high-accuracy data extraction and labeling. • Developed AI workflows supporting structured classification and discrepancy reporting. • Improved efficiency and reduced manual intervention in document labeling and validation tasks.