Close

Abdelrahman Elewah

Resume

About Me

AI Integration Lead and Machine Learning Engineer specializing in designing and deploying scalable AI systems across enterprise and research environments. Currently leading AI and automation integration at MPAC, modernizing property assessment workflows through LLMs, RAG pipelines, and large vision models. Expertise in blueprint understanding, document intelligence, semantic search, and real-time IoT data frameworks. Proven ability to translate research-grade AI into production-ready systems, aligning technical innovation with organizational goals.

Experience

MPAC

Artificial Intelligence Integration Lead

  • Lead enterprise-wide integration of AI and automation solutions across property assessment workflows, modernizing MPAC's digital infrastructure and operational efficiency.
  • Architect and implement scalable data pipelines using Python, LangGraph, FastAPI, Docling, MinerU, PostgreSQL, and MCP servers to support intelligent building plan parsing and document understanding.
  • Design and deploy advanced RAG (Retrieval-Augmented Generation) systems for building plan understanding and semantic search aligned with MPAC assessment standards.
  • Integrate and optimize large vision-language models (Qwen3-VL, LayoutLMv3, PaddleOCR-VL, PP-StructureV3) and detection models such as RF-DETR for architectural object detection, spatial reasoning, and document layout understanding.
  • Lead technical documentation, stakeholder presentations, and cross-team knowledge transfer to ensure sustainable AI adoption.

2U / University of Toronto

Instructional Specialist (part-time)

  • Contributed to the success of the University of Toronto's online Data Analytics Boot Camp, supporting 100+ learners in mastering practical skills for data-driven careers.
  • Facilitated hands-on workshops in Python, Databases, Machine Learning, and Data Visualization, resulting in a 15% increase in student satisfaction scores.
  • Supported the deployment of real-world capstone projects, helping learners apply techniques in domains such as healthcare, HR, and finance.

Ontario Tech University

Research Assistant

  • Designed the SensorsConnect framework, enabling real-time IoT device interoperability and scalable data exchange, inspired by the principles of the World Wide Web. Publication
  • Developed an Agentic IoT Search Engine (ASE-IoT) integrating LLMs, RAG, and autonomous agents for natural language querying of live IoT data. Publication
  • Co-led an OVIN-funded project developing autonomous vehicle curricula and applied research on software-defined vehicles (SDVs) and digital twins.
  • Collaborated with Eagle Aerospace to prototype an Aircraft Deceleration Early Warning System, enhancing runway safety through predictive analytics.

Tamra-IoT

Back-End Developer (part-time)

  • Architected secure and scalable IoT platforms integrating MQTT over TLS, cloud infrastructure, and mobile control interfaces for real-time device management.
  • Designed and deployed Over-The-Air (OTA) firmware update mechanisms, significantly reducing maintenance costs and improving system resilience. Publication
  • Developed a curriculum to teach IoT concepts to high school students, promoting early STEM engagement and hands-on learning.

Education

Ontario Tech University

Jan 2020 – Mar 2025

PhD — Electrical and Computer Engineering

GPA: 4.22/4.3  ·  Official Transcript
Thesis: SensorsConnect: World Wide Web for Internet of Things

Benha University

Feb 2013 – Jan 2018

MSc — Electrical Engineering

Benha University

Sep 2008 – Jun 2012

BSc — Electrical Engineering

Selected Publications

MDPI Sensors  ·  2025

Agentic Search Engine for Real-Time IoT Data

Abdelrahman Elewah, Khalid Elgazzar

View Publication
IEEE Access  ·  Nov 2024

SensorsConnect Framework: World-Wide Web for Internet of Things

Abdelrahman Elewah, Khalid Elgazzar

View Publication
IEEE CEC  ·  Jun 2021

3D-RadViz: Three Dimensional Radial Visualization for Large-Scale Data Visualization

Abdelrahman Elewah, Abeer A Badawi, Haytham Khalil, Shahryar Rahnamayan, Khalid Elgazzar

View Publication

Projects

Localelive — Agentic Search Engine for Real-Time IoT Data

Real-time IoT search engine powered by LLMs and RAG enabling natural language queries across heterogeneous sensor systems. Implemented a semantic search pipeline using Sentence-BERT and HNSW indexing, reducing query latency by 73%. Manages over 37,000 real-time IoT documents across 500+ service types in MongoDB with geo-indexing. Achieved 92% top-1 accuracy in complex intent detection, surpassing Gemini in benchmarks. Deployed on AWS using Docker Compose and Traefik with automated HTTPS and scalable reverse proxy management.

LangGraphFastAPIRAGSentence-BERTMongoDBAWSDocker

Story-to-Movie Recommender Chatbot (RAG-based)

Retrieval-augmented generation (RAG) system combining vector search with LLMs to deliver context-aware movie recommendations, reducing hallucination rates by ~30%. Built a semantic search pipeline enabling retrievals from 1,000+ documents. Applied prompt chaining techniques to improve answer relevance, validated through precision metrics and user feedback. Built with OpenAI API, Pandas, and Tenacity for resilient real-time API usage.

RAGOpenAI APIVector SearchStreamlitPrompt Chaining

Lightweight Fine-Tuning of a Foundation Model (LoRA / GPT-2)

End-to-end NLP pipeline using PyTorch and Hugging Face Transformers with LoRA adapters (PEFT) to fine-tune GPT-2 on the AG News dataset while keeping the base model frozen. Reduced training time and memory usage by over 60% vs. full fine-tuning. Boosted classification accuracy from 83.16% → 88.95%, demonstrating PEFT effectiveness with minimal compute.

PyTorchLoRA / PEFTGPT-2Hugging FaceTensorFlow

RadViz-Plotly — Open-Source Data Visualization Package

Open-source Python package generating interactive 2D and 3D Radial Visualization (RadViz) plots for high-dimensional datasets using Plotly. Enables data scientists and analysts to discover hidden clusters, outliers, and trends through intuitive visual interfaces, improving model explainability and decision-making in analytics workflows.

PythonPlotlyData VisualizationOpen Source

Skills

LLM & Agentic AI

  • LangChain
  • LangGraph
  • LangSmith
  • RAG
  • Agentic AI
  • Docling
  • MinerU
  • MCP

Languages

  • Python
  • C/C++
  • JavaScript

Web & APIs

  • REST APIs
  • FastAPI
  • Flask
  • React
  • HTML
  • CSS

Cloud & DevOps

  • AWS EC2
  • App Runner
  • ECS
  • Docker
  • Kubernetes
  • CI/CD
  • GitHub Actions
  • GitLab CI/CD

Databases

  • PostgreSQL
  • MongoDB
  • MySQL
  • Spark
  • Hadoop

Tools

  • Jenkins
  • VS Code
  • Anaconda
  • Dev Containers

Get in Touch

Have a project, collaboration, or question? I'd love to hear from you.