← Lab

Client work

CODI — document classification with AI

Delivered

An artificial intelligence platform to classify and organise tens of thousands of documents from corporate repositories.

A client needed to bring order to a huge document archive scattered across several repositories (SharePoint, Google Drive, file servers). We built a platform that ingests the documents, classifies them by type and content, and prepares them for structured loading into the target document management system.

The solution combines ingestion connectors, extraction and OCR, a classification engine using language models and entity extraction, deduplication by hash and vector embeddings, and structured export. The taxonomy is configured without writing code. It is sized to process on the order of 10,000 to 100,000 documents.

AIOCREmbeddingsClassificationDocument management

Need something along these lines?

Let’s talk about your project