Omar Daniel Abou Assaf

About

Artificial Intelligence Engineer with a background in software engineering and hands-on experience in Python development, API integration, data processing, and AI-powered application development. Experienced in building scalable solutions with FastAPI, Docker, Git, and modern machine learning workflows.

Experience

• Remote
  Private Tutor
  Feb 2023 – Present
• Coding Expert
  Outlier
  May 2025 – Mar 2026

Education

• Bachelor of Science
  Syria Arab International University
  Computer Science (Artificial Intelligence) · Feb 2019 – May 2024

Skills

• C <3 months
• Deep Learning <3 months
• FastAPI <3 months
• Python <3 months
• Machine Learning <3 months
• Prompt Engineering <3 months
• Streamlit <3 months
• PyTorch <3 months
• TensorFlow <3 months
• LLM evaluation <3 months
• LangChain <3 months
• API Integration Understanding <3 months
• Natural Language Processing <3 months
• Large Language Models <3 months
• LangGraph <3 months
• Fine-tuning <3 months
• Faiss <3 months
• Retrieval Augmentation Generation <3 months
• Generative AI (ChatGPT, Copilot, Gemini) <3 months

Tools / apps / platforms

• ChromaDB <3 months
• Docker <3 months
• Git <3 months

Languages

• Arabic Native / Bilingual Speak · Read · Write
• English Fluent Speak · Read · Write
• German Conversational Speak · Read · Write

Projects

• Autism Spectrum Disorder Classification
  Python, TensorFlow/Keras, Nilearn, PyWavelets

  Developed a deep learning model to classify ASD using rs-fMRI data. Built a neuroimaging pipeline and evaluated the model with 95% accuracy and 96% F1-score.

• Golden Taste Restaurant
  Python, LangChain, Mistral AI, Streamlit

  Built an LLM-powered chatbot using LangChain and Streamlit. Implemented session-based memory for context-aware responses and designed prompt pipelines to improve response quality.

• PDF RAG System
  Python, FastAPI, ChromaDB, LangChain, SentenceTransformer, Mistral AI, Docker

  Built a RAG pipeline for querying PDF documents using LLMs. Implemented chunking, embeddings, and vector-based retrieval, enabling semantic question-answering over PDF data.