Hebrew OCR Specialist for Document AI Enhancement

We are looking for an experienced OCR / Document AI specialist to improve an existing document-processing system.

The system is already developed and operational. The main task is to significantly improve OCR accuracy, structured data extraction, and processing speed, particularly for Hebrew and mixed Hebrew/English business documents.

The documents contain structured information such as:

* Tables
* Item descriptions
* Product/item codes
* Quantities
* Prices
* Document numbers and dates

The specialist will work with an existing developer who will provide access to the current OCR pipeline and assist with integration.

We are specifically looking for someone with proven experience in:

* Hebrew OCR
* RTL documents
* Table extraction
* PDF/image preprocessing
* Structured data extraction
* OCR post-processing and validation
* Improving OCR performance and processing speed
* Document AI / Vision models

Experience with tools/services such as Google Document AI, Azure AI Document Intelligence, AWS Textract, PaddleOCR, Tesseract, OCR/vision models, or similar technologies is relevant, but we are open to the best technical approach.

Important: This is not a project to build an application from scratch. The application already exists. We need a specialist who can analyze the existing extraction pipeline, identify the accuracy/performance bottlenecks, and improve it.

Before hiring, please answer:

1. Have you worked specifically with Hebrew OCR or Hebrew business documents?
2. Have you extracted tables and line items from invoices or similar documents?
3. What OCR/Document AI technologies have you used?
4. How would you measure extraction accuracy before and after your improvements?
5. Can you optimize both accuracy and processing speed?
6. Please provide an example of a similar OCR/document extraction project you have completed.

Additional technical details and sample documents will be provided to shortlisted candidates after initial screening.