I’m Omar Elgemaey, an AI Software Engineer passionate about building high-impact, scalable AI systems. My work spans Generative AI, Large Language Models (LLMs), multi-agent AI systems, computer vision, natural language processing, rare event prediction, and custom AI solutions, bridging the gap between cutting-edge research and practical, real-world applications.

I graduated with a B.Sc. in Electronics and Communications Engineering from Mansoura University and am currently advancing my skills through ITI’s 9-month AI & Machine Learning program, gaining hands-on experience with deep learning, NLP, GenAI technologies, and computer vision pipelines.

I have collaborated with companies such as Corelia, Neqabty, and Cubic information Systems, contributing to AI pipelines, LLM-powered applications, and Generative AI prototypes deployed in real-world environments.

I focus on precision, performance, and clean, reproducible code, ensuring that AI solutions are not only innovative but also scalable, maintainable, and high-performing.

Key strengths: ambition, problem-solving, practical AI development, and a continuous drive to stay at the forefront of AI innovation.

• Information Technology Institute (ITI), MCIT
  9-Month Professional Diploma – Artificial Intelligence & Machine Learning Track
  Topics explored: Generative AI, Transformers, LLMs, RAG systems, NLP, Computer Vision, and Deep Learning pipelines

• Mansoura University
  B.Sc. in Electronics and Communications Engineering – Graduated Very Good with Honors

• Developed AI solutions on local servers for real-life projects.
• Worked on data preprocessing pipelines, model training, and deployment.
• Gained hands-on experience with PyTorch, Transformers, and NLP pipelines.

• Collaborated on projects involving RAG systems, LLM-based information retrieval, and document analysis.
• Built end-to-end pipelines for AI-assisted solutions.
• Optimized models for accuracy, latency, and real-world usability.

• Collaborated with ITI on a hands-on graduation project, leading the design and deployment of AI pipelines for structured and unstructured data.
• Developed and implemented custom ML models and Generative AI prototypes, integrating them into real-world enterprise workflows.
• Gained practical experience in model deployment, Docker containerization, and scalable AI pipelines.
• Focused on advanced NLP and LLM applications, including RAG systems, embeddings, and knowledge-driven AI solutions.

• Python – AI pipelines, GenAI development, and data engineering
• C++ / C – Performance-critical algorithms and low-level programming
• SQL / NoSQL – Data storage, retrieval, and vector embeddings for RAG systems

• Large Language Models: GPT, LLaMA, Claude, MPT
• Transformers: Hugging Face Transformers, Fine-tuning and Inference
• RAG Pipelines: Retrieval-Augmented Generation for knowledge-intensive tasks
• Vector Databases: Pinecone, Weaviate, FAISS
• Prompt Engineering: Designing optimized prompts for multi-turn reasoning
• Agent Systems: Multi-agent pipelines for autonomous decision-making
• LLM Applications: Chatbots, document summarization, code generation, and conversational AI

• OpenCV – Image processing and feature extraction
• YOLO (Ultralytics) – Real-time object detection
• MediaPipe – Pose estimation, hand tracking, and gesture recognition

• Tokenization & Embeddings: Word2Vec, TF-IDF, BERT embeddings
• Libraries: Hugging Face Transformers, spaCy, NLTK
• Tasks: Text classification, summarization, named entity recognition, sentiment analysis

• Web frameworks: FastAPI, Flask, Streamlit
• Containerization & Deployment: Docker, AWS, Azure
• Data Engineering: Pandas, NumPy, MongoDB, SQL, vector databases
• Collaboration: Git, GitHub, JIRA, Agile methodologies

• Built a retrieval-augmented generation system for resume and candidate analysis.
• Integrated vector search with LangChain to handle unstructured documents.
• Achieved high accuracy in candidate matching and recommendation.

• Designed a multi-agent system for task automation and management.
• Leveraged agents powered by LLMs to assign, monitor, and optimize tasks.
• Deployed a real-time monitoring dashboard using Streamlit.

• Developed a conversational AI system for restaurant booking and orders.
• Implemented intent recognition, slot filling, and voice-based interactions.
• Hosted on local servers with multi-modal AI capabilities.

• Built a self-hosted AI email assistant using transformers and LLMs.
• Features included automatic response generation, summarization, and sentiment analysis.
• Integrated with Azure AI services and multilingual voice synthesis.

• Fine-tuning LLMs on domain-specific datasets
• RAG pipeline construction using embeddings and vector similarity
• Text summarization, translation, and question-answering
• Conversational agents and multi-turn dialogue handling
• Code generation & automation using LLMs
• Voice and speech-based AI applications (Arabic & English)
• Evaluation & metrics: BLEU, ROUGE, F1, perplexity, cosine similarity

• Microsoft Azure AI Services – Speech-to-text, TTS, LLM hosting
• AWS / GCP – Model deployment, inference pipelines
• Docker & Kubernetes – Containerized AI pipelines for scalable deployment
• Local Servers – Optimizing AI workflows for on-premise infrastructure

• LinkedIn
• GitHub
• Email