Karim Fayad

Researched and evaluated domain-specific medical LLMs for integration into a clinical decision support system, benchmarking models across accuracy, latency, and clinical relevance metrics. Designed and implemented a knowledge graph pipeline to structure and connect medical knowledge, enhancing LLM response quality and grounding outputs in validated clinical data.

Developed an end-to-end AI auto-fill system for Schengen visa applications with multilingual support. Implemented OCR technology and Large Language Models (LLMs) to automatically extract and populate applicant information from documents, streamlining the visa application process across multiple languages.

Researched and evaluated models for document image classification, selected three optimized for different hardware tiers, and developed a wxPython interface for model and device selection with real-time metrics and preference saving. Implemented a bilingual (English/Arabic) document autofill system using a fast OCR+QA model with RTL Arabic support and automatic form-field mapping, alongside a loca

Developed responsive landing pages with HTML, CSS, and JavaScript, focusing on interactive UI components and performance optimization.

Pursuing a Master of Science in Applied Artificial Intelligence.

Intensive hands-on program covering the modern GenAI engineering stack, from LLM fundamentals, prompt engineering, and RAG pipelines, to AI agents using LangChain, LangGraph, and CrewAI, with applied project builds throughout.

Graduated with a solid background in Computer Science, emphasizing data analysis, machine learning, and web development. Completed coursework in OOP, Data Structures, Databases, and advanced Mathematics.