IDP OCR Engineer

Responsibilities: Build and maintain Python-based OCR and extraction pipelines using queue-based or batch processing. Implement image-quality checks for blur, glare, skew, resolution, and compression to route documents for processing, resubmission, or review. Collaborate with the forensics engineer to distinguish poor image quality from document alteration and reduce false fraud flags. Measure extraction accuracy by document type, calibrate field-level confidence scores against actual error rates, and automate regression testing for every change. Help create clear guidance for users submitting replacement documents.

Qualifications: 3+ years building or operating production document-processing or OCR systems. Hands-on experience with Tesseract, PaddleOCR, Google Document AI, or similar OCR tools. Strong Python skills and experience with image preprocessing, including deskewing, denoising, and contrast adjustment. Experience processing inconsistent inputs and testing extraction accuracy. Bachelor’s or master’s degree in computer science, engineering, or a related field, or equivalent practical experience. Experience with layout-aware models, key-value extraction, image-quality assessment, or eligibility, financial, insurance, or identity documents preferred. Government or regulated-environment experience is a plus.