Omar Ayman Elgemaey

Ai EngineerInternship

• Job Details:1. Technical Track and Specialization I joined the AI Professional Track, focusing on data science, machine learning, and deep learning. The curriculum was structured to provide both depth and breadth across the full AI pipeline. Core Topics Covered: Python for Data Science Mastered data structures, OOP, and libraries like NumPy, Pandas, Matplotlib, and Seaborn. Statistics and Probability Applied concepts like hypothesis testing, distributions, p-values, and confidence intervals. Machine Learning Built supervised and unsupervised models including linear regression, decision trees, random forests, SVMs, K-means, and PCA using scikit-learn. Deep Learning and Neural Networks Used TensorFlow and Keras to build and train CNNs, RNNs, LSTMs, and autoencoders. Natural Language Processing (NLP) Preprocessed text, used TF-IDF and word embeddings, applied models like LSTM and Transformers. Computer Vision Implemented image processing pipelines, applied CNNs, and worked on object detection using YOLOv8. Big Data & Data Engineering Basics Introduction to Spark, Hadoop, and handling large datasets. 2. Soft Skills and Business Acumen The program emphasized more than just technical training. We also had intensive sessions in: Effective communication Presentation and storytelling with data Agile and Scrum methodologies Problem-solving under pressure Time management and working in collaborative teams 3. Graduation Project Our graduation project was a practical application of everything we learned. I worked with a team to build a RAG-based Chatbot for Enterprise Use. Project Highlights: Developed a Retrieval-Augmented Generation (RAG) chatbot using LangChain and Ollama. Integrated FAISS vector database and Sentence Transformers for semantic search. Parsed unstructured documents with PyMuPDF and preprocessed them for retrieval. Built a Streamlit frontend that mimics WhatsApp UI for dual interaction (user and bot). Added fallback and security mechanisms to ensure stable and safe responses. Named the assistant “Zaki” and deployed it on-premises.