Klassif AI

Belgian AI platform specializing in order confirmation processing with human-in-the-loop training and 93% accuracy claims.

klassif-ai

Overview

Founded in 2017 by Tom Vermeulen in Leuven, Belgium, klassif-ai operates as a product of Raccoons, focusing specifically on order confirmation processing rather than general document processing. The company positions against traditional OCR with 93% accuracy versus 60% for legacy solutions and claims 87.3% average reduction in manual effort with 2.5x median ROI over six-month payback periods.

In early 2026, klassif-ai open-sourced its PDF annotation component after three development iterations, demonstrating technical credibility while building developer community engagement. Customer Stijn Hoegaerts at Van Marcke reports "ten hours saved per day on average" through the platform's implementation across manufacturing, logistics, and banking sectors.

The vendor's narrow vertical specialization strategy targets the IDP market's expected 27-30% CAGR growth with continuous learning capabilities via human-in-the-loop training, differentiating from static template-based approaches.

Key Features and Benefits

Order-Specific AI: Specialized algorithms for sales order and invoice processing workflows
Human-in-the-Loop Training: Continuous model improvement through normal employee workflows
Open-Source Annotation Tool: React-based PDF annotation with OCR via Tesseract.js
Model Versioning: Production rollback capabilities with activity monitoring
Enterprise Integrations: Built-in connections to Microsoft Outlook 365, Google Workspace, Dropbox, and ERP systems

Use Cases

Order Confirmation Processing

Manufacturing, logistics, chemicals, food industry, construction, and banking organizations deploy klassif-ai for order confirmation workflows. The platform processes incoming orders via email, PDF, or scanned documents, extracting customer data, product codes, quantities, and pricing with up to 93% productivity increases.

Invoice Management

Organizations process supplier invoices and customer billing documents with 80% faster processing compared to traditional OCR solutions, validating data before ERP system integration.

Technical Specifications

Feature	Specification
Core Technology	Machine learning, NLP with human-in-the-loop training
Document Types	PDF, email, scanned documents
Accuracy Rate	93% vs 60% traditional OCR
Implementation Timeline	4-5 weeks typical deployment
Infrastructure Changes	None required
Parent Company	Raccoons