Document Processing
OCR + RAG to parse invoices and contracts.
Client: B2B Services
Industry: Professional Services
Duration: 5 weeks
Team: 2 engineers
Problem
Hours spent on manual data entry from invoices and contracts.
Solution
OCR + LLM extraction with human-in-the-loop validation and exports to ERP.
Approach
- OCR normalization and layout-aware chunking
- Schema-guided extraction with confidence scoring
- Validation UI and ERP sync with webhooks
Stack
Tesseract/Cloud OCRFastAPIOpenAILangChainPostgreSQL
Results
- Manual entry -70%
- Processing time -60%