Jacob Malick

Built document intelligence systems in Azure using Gen AI and OCR to process unstructured data from 10,000+ grant documents. Developed a Retrieval-Augmented Generation (RAG) pipeline using DBRX, OpenAI, and vector embeddings to extract insights from complex PDF records.

Built document intelligence solutions processing 10,000+ grant documents using Generative AI, OCR, and RAG pipelines to reduce manual processing and optimized Azure data pipelines for high throughput.

Led development of end-to-end ML pipelines in Azure ML Studio for insurance classification models, improving claim processing. Integrated LLM-based APIs (OpenAI, LLaMA) to classify complex document types, reducing misclassification rate.

Architected ML pipelines for insurance claim classification using Azure ML, OpenAI and LLaMA, reducing classification errors and implemented real-time and batch inference with automated drift monitoring.

Designed OCR + object detection solutions using Azure Custom Vision & AFR to process 50k+ prescription documents per month. Improved extraction accuracy from 78% to 95% using hybrid OCR-CV models fine-tuned for domain-specific language.

Designed OCR and computer vision pipelines processing tens of thousands of prescription documents monthly to improve extraction accuracy and automated intake with Azure Logic Apps to cut manual entry.

Developed a recommender system for HR onboarding tasks using collaborative filtering and time-series clustering. Built anomaly detection models to flag high-risk onboarding behaviors using DBSCAN and autoencoders.

Pursued advanced studies in Computer Science, focusing on key areas relevant to modern technological demands. Developed a strong foundation in theoretical and applied aspects of computing.