Boston, Massachusetts, United States
• Primary focus: custom Deep OCR document digitization model built on PyTorch and OpenCV
• Researched and implemented advanced algorithms for recognition and extraction of unstructured data
• Optimized data pre-processing after identifying key trends in customers’ documents, eliminating hours of training time
• Curated extensive training, validation, and testing datasets comprising thousands of internal and external documents
• Implemented robust pipelines to efficiently process model output and seamlessly integrate with other product offerings
• Collaborated cross-functionally to integrate machine learning solutions into product offerings
• Regularly presented research findings, progress updates, and MVP demonstrations to product team, project leads, and C-Suite