Python Developer - Optical Character Recognition

Role Overview

As a Python Developer specializing in OCR and data extraction, you will drive endtoend delivery of scalable documentintelligence solutions. Youll partner with crossfunctional teams to architect and ship robust pipelines that transform unstructured files into actionable data, leveraging cuttingedge OCR libraries and generative AI Responsibilities :

• Architect & Develop : Design and implement Python applications focused on OCR driven data extraction workflows.
• OCR Integration : Integrate and optimize Tesseract, EasyOCR, PaddleOCR and cloudbased services (AWS Textract, Google Vision).
• PreProcessing & Enhancement : Apply OpenCV, PIL and NumPy techniques to clean, segment, and enhance images for superior recognition accuracy.
• Data Structuring : Build custom parsing logic to normalize, validate and store extracted information in downstream systems.
• LLM & Gen AI Enablement : Incorporate Large Language Models (e.g., GPTbased, LayoutLM, Donut) to enrich document understanding, classification, and entity extraction.
• Collaboration & Delivery : Liaise with backend, data science, and DevOps teams to ensure seamless CI/CD integration, monitoring, and support.
• Quality & Documentation : Write modular, test driven code; maintain clear documentation; and conduct peer reviews to uphold code quality and compliance Have Skills :
• Expert proficiency in Python and core automation/data processing libraries (NumPy, Pandas).
• Handson experience with OCR libraries (Tesseract, EasyOCR, PaddleOCR) and handling PDF/image parsing.
• Solid understanding of RESTful API integration and largescale document workflows.
• Proficient in image preprocessing using OpenCV and PIL to maximize OCR accuracy.
• Strong analytical, debugging, and problem solving capabilities in production to Have :
• Proven track record deploying LLM driven document intelligence solutions (e.g., GPT, LLaMA) or Gen AI frameworks.
• Worked with Langchain and Vector DB
• Familiarity with AI document models such as LayoutLM, Donut, TrOCR for advanced layout parsing.
• Exposure to NLP tasks : document classification, entity recognition, semantic search.
• Experience with containerization (Docker), version control (Git), and cloud platforms (AWS/GCP/Azure).
  

(ref:hirist.tech)