AI Document Intelligence Specialist

We are seeking a highly motivated and detail-oriented Document Processing & Automation Specialist with expertise in computer vision, particularly using OpenCV. In this role, you will be instrumental in enhancing the accuracy, reliability, and scalability of our automated document analysis systems. Your contributions will directly impact operational efficiency, data integrity, and the success of our broader digital transformation initiatives. This is a unique opportunity to apply your technical proficiency to real-world challenges and drive meaningful innovation within our organization

Job Responsibilities

• Design, develop, and optimize computer vision algorithms for document classification, segmentation, and data extraction
• Work extensively with OpenCV and other image processing libraries to detect regions of interest, remove noise, and improve OCR readiness
• Conduct performance tuning and error analysis to improve the accuracy and throughput of the document automation system.
• Utilize OpenAI, Azure OpenAI, or similar LLM APIs for intelligent content extraction, reasoning, and validation from OCR text.
• Fine-tune and optimize AI models using prompt engineering, few-shot examples, and structured input/output patterns
• Assist in deploying models to production environments and monitor their performance.

Monitor and analyze system performance, applying continuous improvements to ensure high accuracy and throughput

Must have skills

• Proficiency with OpenCV for tasks like segmentation, denoising, bounding box detection, contour analysis
• Strong understanding of image preprocessing for OCR readiness.
• Python (advanced level) with experience in NumPy, Pandas, and file I/O operations.
• Ability to write clean, modular, and testable code.
• Experience working with at least one OCR engine such as Tesseract, Azure OCR, AWS Textract, or Google Vision
• Hands-on experience with OpenAI APIs (e.g., GPT-3.5, GPT-4) or Azure OpenAI Service.

Prompt engineering skills for structured and unstructured data extraction

Nice to have skills

1.Basic NLP techniques (e.g., entity recognition, regex-based extraction, text cleaning).

2.Experience with tools like spaCy, HuggingFace, or LangChain.

3.Exposure to training or fine-tuning machine learning models (e.g., SVM, CNN, BERT).

4.Experience deploying solutions on Azure, AWS, or GCP