Gen AI-Orchestrator for Email and Document Triage/Routing

About the Project

This project automates the classification and processing of high-volume email requests, especially in domains like commercial banking where customer servicing teams manually read, sort, and respond to hundreds of emails each day. These emails often include attachments, unstructured text, and complex request types that make manual triage slow and error-prone. Our solution uses AI to classify request types, extract sub-requests, process attachments, and detect duplicate emails at scale, reducing operational overhead and improving response efficiency.

Key Features

• Multi-format Attachment Processing: Extracts text from PDFs, DOCX files, images, and scanned documents using OCR to ensure all relevant information is captured.
• Context-based Classification: Uses a retrieval-augmented LLM to identify request types and sub-request types with structured outputs and confidence scores.
• Duplicate Email Detection: Generates embeddings and performs vector searches to identify similar or repeated emails, reducing redundant work.
• Configurable AI Pipeline: Allows users to upload labeled examples and custom prompts so the system can adapt to any classification workflow without retraining.
• Dashboard Interface: Provides a clean UI to upload emails, view classifications, explore similar emails, and adjust model configurations.
• Scalable Architecture: Integrates FastAPI, MongoDB, and an LLM-backed assistant to handle high email volumes efficiently.