Information Extraction Engine
- Extracted information from financial, legal, and business documents using multimodal architectures incorporated with Active Learning Frameworks with accuracy > 80%.
- Designed and deployed an Information Extraction Engine for financial documents, integrating multiple models in a scalable pipeline. Improved document processing speed by > 50%
- Effectively deployed the cohesive ML pipelines to incorporate the well-crafted LLM Pipelines by utilizing models such as OCR, Document Structure and Table Structure Recognition Models existing in silos, deploying the pipeline in a microservice or monolithic architecture as needed.
- Implemented microservices in a Master-Worker Node architecture using Docker, Kubernetes, and AWS.
- Processed legal documents like mortgage and grant deeds using advanced multimodal techniques.
