FR
FR
  • Home
  • About Me
  • Resume
  • Reviews
  • Portfolio
  • Publications
  • Certificates
  • Service
Hello, I'm

Ali Nour Eldin

Applied AI & Software Engineer | GenAI, AI Agents & Intelligent Automation | PhD in AI for BPM

AI Engineer CV (Short Version) Software & AI Engineer CV (Short Version)

Last update: 04/06/2026
About Me

Software & AI Engineer

Software & AI Engineer building production-grade software systems and AI-powered applications.

I'm a Software & AI Engineer with expertise in building production-grade software systems and AI-powered applications. I hold a Ph.D. in AI for Business Process Management (BPM) and have 4+ years of experience spanning software engineering, NLP, LLMs, generative AI, RAG architectures, and AI agents. Skilled in designing scalable systems, integrating AI into real-world products, and delivering end-to-end solutions from research and prototyping to deployment and optimization, I translate business requirements into reliable software solutions with measurable impact.

Resume

Experience, Education & Skills

Professional Experience

  • AI R&D Engineer (Contract)

    Telecom SudParis, France - Hybrid
    September 2025 - Present

    • Designed and implemented a Python-based multi-agent system using LangGraph and Nvidia LLM APIs, leveraging a GraphRAG-powered knowledge graph for cybersecurity question answering and service recommendations.

    • Built an AI pipeline to structure and retrieve cybersecurity assessment questions using embeddings, clustering, and LLMs, reducing processing costs by 40% versus LLM-only approaches.

    • Benchmarked classical ML approaches against LLM-based methods and implemented automated quality evaluation using LLM-as-a-Judge techniques.

    • Mentored PhD students on implementing advanced AI approaches, delivering best practices and hands-on guidance in GenAI.

    • Designed and delivered a process mining workshop featuring a live Celonis demonstration.

    • Supervised 7 engineering students on diverse projects involving LLMs, Retrieval-Augmented Generation (RAG), and Knowledge Graphs (KG).


    Skills: Python (FastAPI), LLMs (OpenAI, Mistral, Anthropic), LangChain, LangGraph, LangSmith, GraphRAG, Docker, Embeddings (OpenAI Embeddings, SBERT), Clustering (sklearn), AWS (EC2).

  • AI R&D Engineer (Contract)

    Siliconcedars, Lebanon - Remote (Parallel Engagement)
    August 2025 - April 2026

    • Defined the end-to-end architecture and built an MVP of an agentic LLM pipeline automating process implementation by generating workflow logic and database schemas.

    • Built the AI agent (generation, validation, human-in-the-loop) with state and memory management plus tool/API calling to execute workflow assembly steps.

    • Implemented the production backend and deployment stack using a Spring Boot and Flowable platform exposing REST APIs for integrations and payment processing, shipped via Docker and GitHub Actions CI/CD on Azure.

    • Presented demos and progress updates to clients and stakeholders and produced technical integration documentation for the internal AI Factory (model deployment, notifications, payment).


    Skills: Agentic workflows, LLMs, LangGraph, LangChain, LangSmith, RAG, human-in-the-loop, Spring Boot, Flowable, REST APIs, Docker, GitHub Actions, Azure, payment integration.

  • R&D Engineer (Contract)

    Bonitasoft, France - Hybrid
    May 2022 - May 2025

    • Architected and built a production-ready Python LLM library (OpenAI, LangChain) for the R&D team, standardizing prompting, guardrails, hallucination mitigation, and evaluation.

    • Delivered multiple low-code/BPM PoCs for AI-assisted process automation and data management, packaging validated prototypes as deployable services (Docker, integration-ready APIs).

    • Implemented an extensible BPMN modeling IDE based on Eclipse GLSP and Java, supporting graphical workflow design, BPMN metamodel manipulation, low-code process engineering, and data-aware workflow modeling.

    • Developed a novel process discovery approach enabling the detection of complex execution patterns not captured by traditional algorithms.

    • Supervised student projects by defining scope, breaking down tasks, setting milestones, and managing time to deliver working prototypes and demos.

    • Produced R&D reporting and supported CIR submissions by documenting experiments, results, and innovation evidence in a compliance-ready format.

    • Developed a novel process discovery algorithm that detects patterns in event logs missed by existing methods while balancing key quality metrics.

    • Presented outcomes in conferences, adapting content for both technical and non-technical audiences.


    Skills: Python, OpenAI, LangChain, LLM evaluation, prompt engineering, guardrails, BPM, BPMN, Process intelligence, Graph theory, low-code automation, Docker, REST APIs, R&D reporting, stakeholder communication.

  • Software Engineer (Freelance)

    MFG Invest, Lebanon - Remote (Parallel Engagement)
    March 2023 - January 2025

    • Built and delivered production fintech platforms, including a MetaTrader5-integrated CRM and a secure E-Wallet, streamlining trading operations and payment workflows.

    • Collaborated with brokers, operations, and compliance stakeholders to gather requirements, translate business needs into user stories, and drive iterative feature delivery.

    • Designed and implemented scalable system architecture, including REST APIs, database schemas, and third-party integrations, with features such as KYC verification, biometric authentication, encrypted transactions, and compliance controls.

    • Produced technical documentation (API references, deployment guides, runbooks) and deployed cloud-native applications with secure infrastructure and real-time notification systems.


    Skills: Laravel, Flutter, REST APIs, MetaTrader5, Fintech Systems, CRM Development, E-Wallets, KYC/Compliance, Cloud Deployment, System Architecture, Project Management.

  • AI R&D Engineer (Part-time)

    Y76, Lebanon - Remote (Parallel Engagement)
    April 2022 - March 2023

    • Automated a DataOps pipeline for face extraction from videos using advanced face detection, cropping, and super-resolution techniques.

    • Fine-tuned generative AI models (StyleGAN2, PyTorch) on curated datasets, achieving high-fidelity cartoon face generation with GCP-based GPU acceleration.

    • Developed and deployed a Flask-based web app for dynamic face generation, optimizing model performance on Google Cloud Platform.


    Skills: Python, PyTorch, StyleGAN2, computer vision, face detection, super-resolution, DataOps, Flask, GCP, GPU acceleration.

  • IT Support Specialist (Contract)

    Hospital, Lebanon - On-site
    April 2020 - March 2022

    • Streamlined hospital IT infrastructure and automated lab data workflows through application development, improving reliability and reducing manual effort.


    Skills: IT infrastructure, automation, application support, Windows Server, VMware ESXi, SQL, MS Access.

  • Data Scientist (Internship)

    Telecom SudParis, France - Remote
    April 2021 - October 2021

    • Created scalable process discovery solutions using Neo4j and Cypher, improving real-time business process analysis.

    • Designed graph-based models for object-centric event data, enhancing data integration and performance over traditional systems.


    Skills: Python (PM4PY), Neo4j (Cypher), Process Mining.

Research Activities

  • Research Project Supervision

    Telecom SudParis, France
    2024 - 2026

    • Project Reliable BPMN Model Updates with LLM Agents (1 student): supervised experimental design, data analysis, and evaluation of agent-based executable business process modeling (2026).

    • Project Auditable Answer Generation for Compliance Using LLMs and Knowledge Graphs (3 students): supervised the experimental protocol, data analysis, and evaluation of generative models designed to satisfy user needs from a knowledge graph (2026).

    • Project Generative Models for Adaptive Questionnaire Recommendation in Cybersecurity (3 students): supervised experimental design, data analysis, and evaluation of generative models for adaptive cybersecurity questionnaire recommendation (2025).

    • Project Leveraging Large Language Models for Business Process Model Extraction from Natural Language (2 students): mentored two third-year students on problem definition, data exploration, and LLM-based prototype development for automatic business process model extraction (2024).

    • Project Bridging the Gap Between Natural Language and UML Class Diagrams Using LLMs (3 students): supervised methodology design, implementation, and evaluation of generative models translating textual descriptions into UML diagrams (2024).

    • Project Advancing Business Data Extraction from Process Descriptions Using Large Language Models (3 students): supervised the development of an LLM-based information extraction pipeline from business process descriptions (2024).

  • Scientific Service

    Committees, journals and conferences
    2025 - 2027

    • Program Committee member:

      - International Conference on Software Engineering (ICSE), Software Engineering Education and Training (SEET) Track (CORE Rank A*), 2027.

    • Conference and journal reviewer:

      - International Journal of Cooperative Information Systems, 2025 - 2026.

      - Information Systems (CORE Rank A*), 2025 - 2026.

      - Sub-reviewer, International Conference on Business Process Management (BPM) (CORE Rank A), 2025 - 2026.

      - International Conference on Cooperative Information Systems (CoopIS), Demo Track, 2025.

      - Sub-reviewer, PhD thesis manuscript, 2025.

  • Research Mentoring

    Telecom SudParis, France
    2026

    • Mentored three PhD students on scientific paper writing, experimental design, and implementation of research prototypes.

  • Teaching Activity

    IMT-TSP, Evry, France
    2024 - 2026

    • Prepared BPM courses for Master 2 computer science students, including practical lab sessions, teaching materials, and hands-on exercises.

Education

  • PhD in Computer Science

    Telecom SudParis, France
    2022 - 2025

    - Thesis title: “A Comprehensive Framework for Low-Code Data-Centric Process Management”

    - Thesis CIFRE collaboration: Telecom SudParis and Bonitasoft

    - Thesis Supervisors: Jonathan Baudot, Dr. Benjamin Dalmas and Prof. Walid Gaaloul

  • Master 2 in Data Science for Risk Analysis

    Faculty of Sciences I, Lebanese University
    2020 - 2021

    - Thesis title: “Enabling Process Mining on Graph Database”

    - Thesis collaboration: Lebanese University and Telecom SudParis

    - Thesis supervisors: Dr. Nous Assy and Prof. Walid Gaaloul

  • Master 1 in Computer Science

    Faculty of Sciences I, Lebanese University
    2019 - 2020

  • Bachelor in Computer Science

    Faculty of Sciences V, Lebanese University
    2016 - 2029

Skills

  • NLP & Gen AI

    Large Language Models (GPT-3.5/4, BERT, T5, Ollama), Retrieval-Augmented Generation (RAG), prompt engineering (few-shot, chain-of-thought), LLM evaluation, guardrails, hallucination mitigation, AI agents, agentic workflows, LangChain, LangGraph, LangSmith, Model Context Protocol (MCP), embeddings, NLTK, spaCy.

  • Machine Learning & Deep Learning

    Supervised and unsupervised learning (classification, k-NN, k-means, PCA, decision trees, naive Bayes, HMM), deep learning (CNN, RNN, LSTM, GANs), transfer learning, graph theory, process mining, PyTorch, TensorFlow, scikit-learn, MLflow, OpenCV.

  • Software Engineering

    Python (FastAPI, Flask), Java (Spring Boot), REST APIs, PHP (Laravel), Dart (Flutter), Vue.js.

  • DevOps & Cloud

    Docker, Git, CI/CD (GitHub Actions), Azure, AWS, JFrog, GCP.

  • Databases

    SQL, NoSQL, graph databases, Firebase, vector databases.

Recommendations

What collaborators say

I highly recommend Ali. We worked closely together on the GovX Project, where he consistently demonstrated strong planning abilities and flawless execution. Ali is a delivery-oriented professional who takes complete ownership of his work, ensuring high-quality results are met on time. Beyond his dedication and technical focus, he is an incredibly positive and easy person to collaborate with, making him a fantastic asset to any team.

Imad El Anis

Senior Operation Manager at SiliconCedars Managed Ali directly Recommended on LinkedIn, June 1, 2026
LinkedIn recommendation

I had the opportunity to supervise and work closely with Ali over several years in both research and collaborative project environments. Throughout this period, Ali consistently demonstrated a strong balance between research thinking and engineering execution.

He has a strong ability to explore complex topics, quickly structure new ideas, and identify practical paths toward concrete solutions. Beyond his technical skills, Ali stands out for his autonomy, analytical mindset, and capacity to plan and drive work efficiently while maintaining high standards and respecting deadlines.

Ali is curious, proactive, and highly committed in everything he undertakes. It was always a pleasure to exchange ideas and collaborate with him, and I am confident he would bring strong value to any research or engineering team.

Walid Gaaloul

Prof.dr.ir. Managed Ali directly Recommended on LinkedIn, May 14, 2026
LinkedIn recommendation

I had the opportunity to work with Ali on collaborative AI and LLM-related projects. He demonstrated strong expertise in prototyping, implementation, benchmarking, and evaluation of different AI approaches and LLMs depending on the targeted use cases and technical constraints.

Ali also actively contributed to the coordination of work packages within collaborative projects, combining solid research skills with a strong engineering mindset to transform ideas into concrete and functional solutions. He is autonomous, rigorous, collaborative, and always focused on delivering practical and effective results.

I am confident Ali would be a strong asset to any research or engineering team.

Mohamed Sellami

Associate Professor at Télécom SudParis Managed Ali directly Recommended on LinkedIn, May 13, 2026
LinkedIn recommendation

I had the pleasure of working with Ali Nour Eldin, and he stands out as one of the most dedicated, driven, and organized professionals I've collaborated with. He brings together strong technical expertise with the ability to quickly adapt, learn, and effectively tackle complex challenges.

Ali is also a dependable and talented researcher who approaches problems with curiosity, persistence, and a solution-oriented mindset. We had the opportunity to work together on an AI digitalization assistant that transforms documented services into fully deployed services backed by BPMN workflows and dynamic UI generation. His contributions played an important role in turning complex ideas into scalable and practical solutions.

Beyond his technical strengths, Ali is reliable, collaborative, and always committed to delivering high-quality work. I would gladly work with him again in the future, whether on a research initiative or an engineering project.

Mohammad Osta

Research And Development Manager at SiliconCedars Managed Ali directly Recommended on LinkedIn, May 12, 2026
LinkedIn recommendation

Ali is one of the most dedicated and hard-working people I have had the privilege of working with. He perfectly combines his broad technical knowledge, advices of his peers and his intuition to maximize the value of an idea or a project. I would hire him again in a heartbeat if I had the opportunity, whatever a research or engineering topic.

Jonathan Baudot

Site Intelligence | Low-Power & Autonomous Vision Expert Managed Ali directly Recommended on LinkedIn, November 10, 2025
LinkedIn recommendation
Portfolio

My works that I did

  • TPRA Question Answering

    In the TPRA process, suppliers receive security questionnaires from different organizations. Even when the questions are based on the same cybersecurity standards, they are written differently and still need to be answered using the supplier's existing security documentation.

    The main limitation is the time consumed by manually reading documents, finding the relevant evidence, and rewriting answers for each questionnaire. This becomes harder when the documentation is incomplete, ambiguous, contradictory, or spread across many files.

    I implemented a rerank-based retrieval method to improve the selection of evidence before sending it to the LLM. The pipeline combines vector retrieval with reranking so the model receives more relevant document chunks and can generate answers grounded in the existing documentation.

    I am also optimizing this approach with GraphRAG. I designed a multi-agent GraphRAG solution with a schema agent, an extraction agent, and a knowledge-graph QA agent. The system extracts entities from the documents, enriches them, builds relations, retrieves relevant quotes, and uses those references to answer each TPRA slot. The backend combines Neo4j for the knowledge graph, ChromaDB for vector retrieval, and LangChain/LangGraph for LLM orchestration and evaluation.


    Technologies: Python, LangChain, LangGraph, LangSmith, Neo4j, ChromaDB, Docker, OpenAI/NVIDIA/Groq models, GraphRAG, hybrid retrieval, cybersecurity QA.

  • GovX - Digital Government Services

    GovX is a Lebanon Works initiative focused on building a national coordination hub for digital governance, smarter public services, and stronger institutions. The project brings together public-sector actors, private-sector collaborators, and technology partners to accelerate AI-enabled government service delivery in Lebanon.

    My contribution was delivered through SiliconCedars, one of the project collaborators. I worked on the technical foundation for an AI Factory concept that can help design, generate, validate, and deploy digital government services from structured process requirements.

    I developed an agentic LLM pipeline that translates process descriptions into workflow logic and database schema proposals. The system includes generation, validation, human-in-the-loop review, state and memory management, and tool/API calling so the agent can assemble implementation steps instead of only producing static text.

    I also built the backend and deployment stack around Spring Boot and Flowable, exposing REST APIs for service orchestration, integrations, notifications, and payment-related flows. The delivery setup used Docker, GitHub Actions CI/CD, and Azure deployment practices, with technical documentation and demos for stakeholders.


    Technologies: LangGraph, LangChain, LLMs, agentic workflows, human-in-the-loop validation, Spring Boot, Flowable, REST APIs, Docker, GitHub Actions, Azure, service orchestration, AI Factory.

    Reference: GovX on Lebanon Works

  • TPRA Question Generation

    In third-party risk assessment, organizations need to evaluate the cybersecurity posture of suppliers that provide services, handle sensitive data, or support critical systems. The questions are usually derived from standards such as ISO, NIST, NIS2, and SOC.

    The limitation is that question generation is not a simple one-to-one mapping from a standard to a questionnaire. Each supplier has its own services, tools, responsibilities, and risk exposure. Generic questions miss important context, while manual adaptation is time-consuming and does not provide a standard way to ask or generate consistent questions.

    I worked on an AI pipeline that organizes the question repository with reusable labels before generating or selecting questions. Instead of relying only on semantic search over raw question text, the system retrieves questions through a label-based layer that captures control domain, assessment intent, and evaluation scope. This makes the retrieval more predictable for cybersecurity teams and easier to audit.

    The engineering design uses LLMs selectively to create and normalize labels, then reuses those labels across the repository with clustering and kNN-based propagation. This reduces repeated LLM calls, lowers inference cost, and keeps question selection consistent when the repository grows. The generated questionnaires are supplier-aware, aligned with the original controls, and easier to maintain as a product feature.


    Technologies: Python, Large Language Models (LLMs), label-based retrieval, semantic labeling, embedding clustering, kNN label propagation, prompt engineering, cybersecurity standards, TPRA/TPRM.

  • Weather Telegram Bot

    I built a Telegram bot that provides real-time weather updates using LangGraph and the Model Context Protocol (MCP). The bot interprets user queries, fetches data from a weather API, and responds naturally in chat.

    As part of the project, I tested two agent types: a ReAct-style agent, which reasons step by step while calling tools, and a stateful graph agent, which manages multi-turn conversations with context memory. This comparison highlights different approaches to designing conversational AI flows.


    Technologies: Python, LangGraph, MCP, Telegram Bot API, Weather API, LLMs.

    GitHub: https://github.com/NourEldin-Ali/mcp-langgraph-weather-telegram

  • RAG-QA Chatbot

    For this project, I focused on building a context-aware Retrieval-Augmented Generation (RAG) chatbot using modern AI frameworks and tools. Leveraging Large Language Models (LLMs), vector databases, and retrieval pipelines, the chatbot delivers accurate and contextually relevant answers from unstructured data sources. The system is architected using Python, LangChain, FastAPI, and Streamlit, combining LLM-driven natural language processing with real-time user interaction.

    I used LangChain to orchestrate the flow between document retrieval and LLM prompting. FastAPI serves as the backend API layer, managing interactions between the user interface, the vector database (ChromaDB), and the LLMs. By embedding documents into vector space using LangChain’s retrieval pipeline and storing them in ChromaDB, the system performs fast and efficient similarity searches based on user queries. This architecture ensures responses are grounded in retrieved content, increasing both accuracy and reliability.

    On the frontend, Streamlit was used to develop a clean, user-friendly interface for real-time question answering. Users can input natural language queries, which are then processed through the backend pipeline. Retrieved documents are passed to the LLM, which synthesizes a response based on the content. The system is compatible with both OpenAI models and local LLMs via Ollama, supporting models like Mistral and LLaMA for flexible deployment.

    This project demonstrates my ability to integrate retrieval mechanisms with generative models, handle vector database operations, and build end-to-end applications that combine multiple AI components. It highlights my skills in prompt engineering, API development, and full-stack LLM applications. The modular and scalable design allows easy adaptation for enterprise-level knowledge assistants or internal document Q&A tools.


    Technologies: Python, FastAPI, Streamlit, LangChain, ChromaDB, Ollama, OpenAI, Large Language Models (LLMs).

    GitHub: https://github.com/NourEldin-Ali/RAG-QA-Chatbot

  • Natural Language to BPMN/UML/Business Data using LLM

    For this project, I focused on leveraging cutting-edge technologies like Large Language Models (LLMs) and implementing prompt engineering techniques to automate business process modeling. Using Python, Flask, and Langchain, I built a hybrid system that integrates LLM-based text analysis with a structured algorithmic approach. This approach decomposes the complex task of generating business process models from textual descriptions into manageable sub-tasks. The LLM analyzes, clarifies, and extracts process entities such as activities and dependencies, while the modular pipeline provides users with an interactive platform for validating or updating extracted information, resulting in highly accurate BPMN models.

    Additionally, I utilized the power of Python and Langchain to integrate the LLM's natural language understanding capabilities within a scalable web framework developed using Flask. This hybrid system not only improves efficiency by reducing token usage but also enhances the accuracy and relevance of generated models. By breaking down the process modeling into distinct steps, this project showcases my skills in prompt engineering, Python development, and the use of AI technologies to solve real-world problems, offering an innovative approach to handling incomplete or non-expert descriptions in process automation.

    Furthermore, the project extends its capabilities to support the transformation of textual descriptions into Unified Modeling Language (UML) diagrams. The system extracts structural and behavioral elements from the text, such as classes, attributes, methods, and their relationships, to automatically generate accurate UML diagrams, including class diagrams and sequence diagrams. This functionality aids in visualizing complex software architectures and enhances the clarity of system designs derived from natural language inputs.

    In addition to BPMN and UML generation, the system is designed to transform textual data into business data models that can be directly used to execute BPMN workflows. By mapping the extracted entities and process details into executable business data formats, the project facilitates seamless integration with BPMN engines, thereby enabling automated process execution. This capability ensures that the generated models are not only descriptive but also operational, bridging the gap between process design and execution.


    Technologies: Python, Flask, Langchain, Large Language Models (LLMs), BPMN, nodejs (Vite Framework), AWS, plantUML.

  • E-Wallet System

    This E-Wallet system is a comprehensive digital payment solution comprising an Admin Portal, a Mobile App, and a Backend developed using Laravel. It supports money transfers, currency exchanges, product purchases, and secure user authentication via KYC and OTP mechanisms.


    1. Admin Portal (Flutter Windows)

    Admins can manage users, transactions, currencies, trading groups, and system settings, ensuring compliance with regulations. Key features include transaction management, fee handling, KYC approval, and integration with payment gateways and Firebase Cloud Messaging for real-time notifications.


    2. Mobile App for Agents & Clients (Flutter Mobile)

    Clients and agents can manage their accounts, perform internal transfers, request withdrawals, deposit via credit card, and create real or demo accounts. Biometric authentication, QR code payments, and OTP provide enhanced security.


    3. Backend (Laravel)

    The Laravel backend manages user data, KYC verification, OTP generation, and secure transaction processing. It integrates real-time currency exchange, product and merchant management, and ensures regulatory compliance through encryption and robust security protocols.


    This E-Wallet system is a robust solution designed to handle digital payments, currency exchanges, product purchases, and secure onboarding through KYC and OTP, ensuring high levels of security and regulatory compliance for users, agents, and administrators.


    Technologies: Flutter, Laravel, Firebase Cloud Messaging, MySQL.

  • CRM System

    CRM System for Trading Account Management

    This CRM system is designed for managing trading accounts and facilitating financial deposits and withdrawals, directly integrated with MetaTrader5 via API. It provides tools for both clients and brokers to manage their accounts and transactions securely and efficiently. The backend is powered by Laravel for robust data management and security.

    • Admin Portal (Flutter Windows): Manage users, currencies, transactions, trading groups, and client trading accounts, with full synchronization to MetaTrader5.
    • Mobile App (Flutter): Clients and brokers can manage trading accounts, perform internal transfers, request cash-out, and deposit funds via credit cards. Brokers can additionally manage clients and deposit to client accounts.
    • MetaTrader5 Integration: Real-time synchronization of trading accounts, automated deposits, withdrawals, and account management.
    • Backend (Laravel): The backend is built with Laravel, ensuring secure handling of user data, transactions, and API communication with MetaTrader5.

    Technologies: Flutter, Laravel, MetaTrader5 API, MySQL.

  • Business Static Website for Innovatech

    I developed this website for Innovatech. It is a clean, responsive static site designed to showcase the company's services and brand identity. The website incorporates modern web technologies and ensures cross-platform compatibility, providing an optimal user experience on both desktop and mobile devices.


    Website: https://innova-tech.me

    Technologies: HTML5, CSS, JavaScript, PHP (for email functionality).

  • Workflow Web Modeler

    This project is an open-source framework designed to facilitate the modeling and manipulation of DF-BPMN compliant workflows which is an extension of BPMN 2.0. Built on the Eclipse Graphical Language Server Platform (GLSP), it offers a flexible Java-based metamodel for generating, importing, and exporting BPMN models, as well as supporting the extension of BPMN elements. This project includes both a client-side graphical modeling tool and a server-side component that allows for seamless integration with modern web technologies. It's an ideal tool for developers looking to build or extend BPMN 2.0 workflows within custom applications, making it highly adaptable for business process automation scenarios​


    Technologies: Java (Eclipse GLSP), BPMN, Typescript

    Source Code: https://github.com/NourEldin-Ali/open-bpmn

  • Generate Cartoon Faces using StyleGan

    I fine-tuned StyleGAN to generate high-quality cartoon faces. Using a custom dataset, I adapted the model to produce diverse cartoon images with distinct features. Moreover, I modified the original StyleGAN architecture to incorporate conditional inputs, allowing precise control over the generated faces. This enhancement enables the generation of cartoon faces with specific attributes, such as gender (male or female), age (young or old), and skin tone. These customizations were achieved by manipulating the latent space and refining the model to respond to these conditions, ensuring the output meets the desired criteria.

    Indeed, I modified facial emotions in generated images by manipulating the latent space directions of StyleGAN. I explored specific vectors in the latent space that correspond to various emotions like happiness, sadness, anger, and surprise. By adjusting the direction of these vectors, I was able to seamlessly change the emotional expression of generated faces while maintaining their core identity and other attributes, such as gender and age. This approach allows for precise and dynamic control over facial emotions in generated images, demonstrating the power of StyleGAN's latent space manipulation.


    Technologies: Python, Pytorch, StyleGAN2, Google Colab, Google Cloud Platform.

  • Face Recognition

    I developed a workflow to automate the process of extracting and enhancing cartoon faces from video files. The workflow begins by downloading cartoon movies, followed by detecting all faces in the video frames. Using advanced face detection algorithms, the process specifically targets cartoon faces, ensuring they are aligned to a specific position. Each face is then cropped to a predefined resolution. Additionally, I integrated a super-resolution model to upscale and enhance the quality of the extracted faces, ensuring high-resolution output. This end-to-end pipeline efficiently processes cartoon movies, resulting in high-quality, precisely cropped facial images suitable for further applications.


    Technologies: Python, Google Cloud Platform, Super Resolution Model.

  • Face Transformation using Semantic StyleGAN

    I employed Semantic StyleGAN to perform targeted facial transformations by modifying specific facial segments. Leveraging semantic manipulation in the latent space, I was able to alter individual features such as hair style, hair color, and add facial hair (e.g., beards or mustaches) without affecting other aspects of the face. By isolating and adjusting these distinct facial regions, the transformations maintained a natural and cohesive appearance. This approach enables high flexibility in customizing specific facial attributes while preserving the overall identity of the generated face.


    Technologies: Python, Pytorch, Semantic StyleGAN, Google Colab, Google Cloud Platform.

  • Incubator hand made using Arduino

    I developed a fully functional handmade incubator using an Arduino microcontroller. The primary goal of this project was to create a system that maintains optimal conditions for egg incubation, specifically controlling temperature and humidity, while incorporating an automatic egg-turning mechanism. This incubator offers a low-cost, DIY solution that mimics the environmental conditions of a traditional incubator, making it ideal for hobbyists and small-scale poultry breeders.


    Components: Arduino Uno, DHT22 Senso, Servo Motor, LCD Display, Relay Module, Power Supply.

  • Blood Donation System

    The LifeSaver mobile application is a cross-platform tool designed to simplify the process of blood donation, making it more accessible and efficient. Built using Flutter, an open-source framework developed by Google, the app allows users to register as blood donors or request blood donations through a seamless interface available on both Android and iOS. Flutter, using the Dart programming language, was selected for its efficiency in creating high-performance apps with a single codebase for both platforms. The integration of Flutter ensures that the application delivers a smooth and native-like experience while maintaining consistency across devices. ​

    The LifeSaver app addresses the challenge of timely communication between hospitals and potential blood donors. It connects users to hospitals in need of blood donations and allows them to respond to requests, track their donation history, and earn points based on their contributions. The app also features real-time notifications, allowing hospitals to post urgent requests and donors to be alerted instantly based on their proximity to the healthcare centers. This system reduces the delays often experienced in traditional blood donation processes, making the app a critical tool in saving lives​


    Technologies: Flutter, Dart, Firebase Cloud Messanging.

    Source Code: https://github.com/NourEldin-Ali/Blood-Donation

  • School Management System

    I developed an Android application titled "School Student Management." The project aimed to enhance the interaction between students, teachers, and administrators by streamlining class management and communication. The application allows teachers to assign tasks, students to view and complete their assignments, and administrators to manage user accounts, classes, and courses. The technical backbone of the project involved using Java for coding, XML for design, and Firebase for database management, messaging, and notifications. This solution simplifies the educational process by creating an efficient and user-friendly interface for both students and teachers, while also providing administrative control over key educational functions. ​


    Technologies: Android Java, Firebase Cloud Messanging, Firebase.

Publications

Research Publications

Certificates

Courses & Certificates

  • Natural Language Processing Specialization

    Date: February 2025

    Natural Language Processing Specialization

    I have successfully completed all four courses in the Natural Language Processing Specialization by DeepLearning.AI on Coursera.

    Completed Courses:

    • Natural Language Processing with Classification and Vector Spaces: Built foundational understanding of NLP tasks such as text classification and sentiment analysis. Learned to represent text using bag-of-words, TF-IDF, and word embeddings. Implemented Naive Bayes and logistic regression classifiers for real-world NLP applications.
    • Natural Language Processing with Probabilistic Models: Gained expertise in sequence models for NLP, including Hidden Markov Models and the Viterbi algorithm. Applied these models to tasks such as part-of-speech tagging and named entity recognition using probabilistic methods.
    • Natural Language Processing with Sequence Models: Learned how to build and train recurrent neural networks (RNNs), GRUs, and LSTMs for NLP tasks. Applied sequence models to machine translation, text generation, and spelling correction.
    • Natural Language Processing with Attention Models: Mastered attention mechanisms and transformer models, including the foundational concepts behind BERT and GPT. Implemented attention-based architectures for question answering, translation, and summarization tasks.

    Certificate Link: https://coursera.org/verify/professional-cert/YKL2MJTK0BEK

  • MLOps | Machine Learning Operations

    Date: February 2025

    MLOps | Machine Learning Operations

    I have successfully completed all four courses in the MLOps | Machine Learning Operations Specialization by Duke University on Coursera.

    Completed Courses:

    • Python Essentials for MLOps: Developed fundamental Python programming skills, including working with variables, data types, functions, and classes. Gained proficiency in using libraries such as Pandas and NumPy for data manipulation and analysis, essential for MLOps workflows.
    • DevOps, DataOps, MLOps: Explored the principles and practices of DevOps, DataOps, and MLOps. Learned to build operational pipelines, implement continuous integration and delivery, and deploy machine learning models in production environments.
    • MLOps Platforms: Amazon SageMaker and Azure ML: Gained hands-on experience with leading cloud-based MLOps platforms. Applied exploratory data analysis techniques, built and deployed machine learning models using AWS SageMaker and Azure Machine Learning, and understood the nuances of cloud-based MLOps solutions.
    • MLOps Tools: MLflow and Hugging Face: Mastered the use of prominent open-source MLOps tools. Created and managed MLflow projects for model tracking and lifecycle management, utilized Hugging Face models and datasets to build APIs, and deployed containerized models to the cloud.

    Certificate Link: https://coursera.org/verify/professional-cert/SHSDCYDORDWF

  • PyTorch Ultimate 2024 - From Basics to Cutting-Edge

    Date: January 2025

    PyTorch Ultimate 2024

    I have successfully completed all three courses in the PyTorch Ultimate 2024 - From Basics to Cutting-Edge Specialization by Packt on Coursera.

    Completed Courses:

    • Foundations and Core Concepts of PyTorch: Established a solid understanding of PyTorch, including setting up the environment, grasping fundamental AI and machine learning concepts, and building, training, and evaluating neural networks from scratch.
    • Building and Training Neural Networks with PyTorch: Developed skills in constructing and training neural networks for various tasks, such as implementing classification models with multi-class and multi-label datasets, utilizing Convolutional Neural Networks (CNNs) for image and audio classification, and applying object detection techniques using the YOLO algorithm.
    • Advanced PyTorch Techniques and Applications: Explored advanced topics, including designing autoencoders for dimensionality reduction, building Generative Adversarial Networks (GANs) for data simulation, developing Graph Neural Networks (GNNs) for graph data, implementing Transformers and Vision Transformers, and deploying models using Flask on Google Cloud.

    Certificate Link: https://coursera.org/verify/professional-cert/E2EMBY87JKTC

  • Machine Learning Specialization

    Date: Decembre 2024

    Machine Learning Specialization

    I have successfully completed all three courses in the Machine Learning Specialization by Andrew Ng (Coursera).

    Completed Courses:

    • Supervised Machine Learning: Regression and Classification: Gained a strong understanding of foundational supervised learning techniques, including linear and logistic regression, evaluation metrics, and essential preprocessing steps.
    • Advanced Learning Algorithms: Explored advanced supervised learning techniques, including neural networks, decision trees, ensemble methods, and insights into tradeoffs between bias and variance.
    • Unsupervised Learning, Recommenders, Reinforcement Learning: Learned about clustering, principal component analysis (PCA), recommendation systems, and the basics of reinforcement learning.

    Certificate Link: https://coursera.org/verify/professional-cert/E2EMBY87JKTC

  • Deep Learning Specialization

    Date: Novembre 2024

    Deep Learning Specialization

    I have successfully completed all five courses in the Deep Learning Specialization by Andrew Ng (Coursera).

    Completed Courses:

    • Neural Networks and Deep Learning: Developed a solid foundation in building and training neural networks, focusing on core concepts like backpropagation, activation functions, and deep neural networks.
    • Improving Deep Neural Networks: Gained expertise in hyperparameter tuning, regularization techniques, and optimization algorithms to enhance model performance and efficiency.
    • Structuring Machine Learning Projects: Learned best practices for diagnosing and improving machine learning models, managing overfitting/underfitting, and structuring scalable projects.
    • Convolutional Neural Networks: Mastered techniques for building and training convolutional neural networks (CNNs), including applications in computer vision tasks such as image recognition, object detection, and transfer learning.
    • Sequence Models: Acquired knowledge of recurrent neural networks (RNNs), long short-term memory networks (LSTMs), and attention mechanisms for tasks such as natural language processing, and speech recognition.

    Certificate Link: https://coursera.org/verify/professional-cert/2IFHVU55O343

  • Prompt Engineering Specialization

    Date: January 2024

    Prompt Engineering Specialization

    I completed a three-course specialization in Prompt Engineering offered by École Polytechnique through Coursera. The courses focused on mastering techniques for crafting prompts that leverage generative AI tools like ChatGPT. I learned key techniques including advanced prompt patterns, automating tasks like data analysis, generating personalized content, and solving problems with AI-driven solutions. The hands-on projects enhanced my ability to apply these models in real-world contexts, from automating processes to generating creative solutions. This experience has significantly boosted my AI literacy and productivity skills.

    Courses:

    • Prompt Engineering for ChatGPT
    • ChatGPT Advanced Data Analysis
    • Trustworthy Generative AI

    Certificate Link: https://coursera.org/verify/specialization/ELSPHXD3BWPF

  • Google Project Management: Professional Certificate

    Date: January 2024

    Google Project Management Certificate

    I have successfully completed 3 out of 6 courses in the Google Project Management Certificate on Coursera. These courses provided me with a strong understanding of the key phases in project management and equipped me with practical skills to effectively lead and manage projects.

    Completed Courses:

    • Foundations of Project Management: Gained insights into essential concepts, frameworks, and methodologies, building a solid foundation for planning and executing projects successfully.
    • Project Initiation: Starting a Successful Project: Developed practical skills in defining project goals, managing stakeholders, and initiating projects with clarity and effectiveness.
    • Project Planning: Putting It All Together: Learned to create detailed project plans, including timeline setting, resource management, and risk mitigation, ensuring project success.

    While I have not yet completed the remaining three courses, the knowledge and skills acquired from the completed courses have provided a strong foundation in project management principles.

  • Google IT Support Professional Certificate

    Date: March 2022

    Google IT Support Professional Certificate

    I have successfully completed the Google IT Support Professional Certificate, a program consisting of five comprehensive courses designed to build expertise in IT support. Through this program, I developed hands-on skills in troubleshooting, system administration, networking, operating systems, security, and customer support, preparing me for roles in technical support and IT operations.

    Completed Courses:

    • Technical Support Fundamentals: Gained foundational knowledge in IT support, including an overview of computer hardware, software, troubleshooting, and customer service best practices.
    • The Bits and Bytes of Computer Networking: Learned the fundamentals of networking, including protocols, networking services, troubleshooting, and the basics of cloud computing.
    • Operating Systems and You: Becoming a Power User: Acquired skills in navigating and managing operating systems, with a focus on Linux and Windows environments, command-line tools, and system optimization.
    • System Administration and IT Infrastructure Services: Explored system administration tasks such as configuring servers, managing services, and deploying infrastructure for organizational IT needs.
    • IT Security: Defense Against the Digital Dark Arts: Developed an understanding of security concepts, including encryption, access control, network security, and incident response strategies.

    Certificate Link: https://coursera.org/verify/professional-cert/W2QKPEQMYBSR

  • Neo4j Graph Database

    Date: August 2021

    Neo4j Graph Database

    I have completed a comprehensive set of six courses through the Neo4j Academy, gaining hands-on experience with graph databases and Cypher query language. These courses include topics such as querying with Cypher, creating nodes and relationships, using indexes for query optimization, and best practices in Neo4j 4.x. Additionally, I gained skills in importing data into Neo4j and designing graph data models tailored to specific use cases. My training provides me with a solid foundation to design, implement, and optimize graph-based applications efficiently using Neo4j.


    Certificates Number:

    • Overview of Neo4j 4.x: 49278841
    • Querying with Cypher in Neo4j 4.x: 507560
    • Creating Nodes and Relationships in Neo4j 4.x: 36106601
    • Using Indexes and Query Best Practices in Neo4j 4.x: 96999171
    • Importing Data with Neo4j 4.x: 20600868
    • Graph Data Modeling for Neo4j: 67161317

  • Routing and Switching: Introduction to Networks

    Date: July 2019

    Routing and Switching: Introduction to Networks

    I have successfully completed the Cisco CCNA1 Routing and Switching: Introduction to Networks course, conducted onsite at the Lebanese University. This introductory program provided me with a solid foundation in basic networking concepts, including network architecture, protocols, and services. I gained hands-on experience in configuring routers and switches, as well as troubleshooting basic network issues. This course introduced me to essential LAN and WAN technologies, preparing me for further studies in networking and modern network infrastructures.

Service

Never compromise with quality

  • AI Consulting and Development

    I specialize in delivering innovative AI solutions that help businesses stay ahead of the curve.

    With years of experience in AI technologies, I specialize in delivering innovative AI solutions that help businesses stay ahead of the curve. My hands-on experience with cutting-edge technologies like Generative Adversarial Networks (GANs), StyleGANs, and Large Language Models (LLMs) enables me to provide advanced AI consulting and development services tailored to your needs. I’ve worked on projects where LLMs were employed to convert text into Business Process Model and Notation (BPMN), transforming complex business processes into efficient, automated workflows. My deep technical knowledge allows me to create AI-powered solutions that automate tasks, improve decision-making, and deliver tangible business outcomes.

    I Having worked across various industries, I understand the intricacies of applying AI in real-world contexts. Whether it’s automating repetitive tasks or enhancing data-driven decision-making, my solutions are designed to seamlessly integrate into your operations. Every AI project I undertake is rooted in my passion for innovation and my commitment to delivering results that drive real value for your business.

    Key AI Projects:

    • Developed AI models using GANs and StyleGANs for creative image generation, elevating marketing strategies for businesses.
    • Utilized LLMs to transform textual descriptions into BPMN diagrams, significantly optimizing business processes.

  • Software Developement

    I bring a wealth of experience in software development, having led and executed various projects across multiple platforms.

    I bring a wealth of experience in software development, having led and executed various projects across multiple platforms. My expertise spans mobile apps, desktop applications, and enterprise software, giving me a comprehensive understanding of what it takes to create functional, scalable, and user-friendly solutions. From developing a highly secure E-wallet app to building a comprehensive CRM system, my experience enables me to tackle complex challenges with ease. I’ve also developed a blood donation mobile app, a school management system, and a client archiving solution, showcasing my versatility and ability to deliver tailored software solutions across different sectors.

    With every project, I focus on understanding the unique needs of the business to create custom software that drives efficiency, enhances user engagement, and supports business growth. My hands-on approach ensures that each solution is meticulously designed, rigorously tested, and deployed with precision. When you work with me, you’re not just getting a developer—you’re partnering with someone who is dedicated to delivering a product that exceeds your expectations and helps your business succeed.

    Key Software Development Projects:

    • Built a secure and user-friendly E-wallet mobile app for a financial institution, streamlining transactions.
    • Developed a CRM system that improved customer interaction and data management for a service-based company.
    • Designed a blood donation mobile app to bridge the gap between donors and healthcare centers, improving accessibility.
    • Created a school management system to simplify administrative workflows for educational institutions.
    • Delivered a client archiving system to enhance data organization and retrieval for professional service firms.

  • IT consultation

    I offer expert IT consultation services that optimize and enhance your systems

    With a rich background in IT infrastructure and cloud computing, I offer a full range of IT Consultation services designed to optimize your technology stack and ensure seamless operations. My expertise spans traditional networking environments like ESXi and Windows Server, cloud platforms such as AWS and Google Cloud Platform (GCP), and hosting solutions for web applications. Whether you need help managing on-premise systems, migrating to the cloud, or optimizing your cloud infrastructure, I provide tailored solutions that ensure your IT systems are secure, scalable, and efficient.

    I have successfully integrated complex IT systems, enabling seamless communication between different technologies and platforms. For example, I have integrated hospital lab machines with hospital management systems, automating data entry and enhancing accuracy. Additionally, I specialize in troubleshooting and resolving IT issues to minimize downtime and ensure smooth business operations. Whether it’s setting up secure hosting environments, optimizing network infrastructure, or connecting systems to ensure smooth data flow, I work to ensure that your IT systems are not only optimized for today but also prepared for future growth.

    Key IT Projects:

    • Managed and optimized virtualized environments using ESXi and Windows Server, improving operational efficiency for enterprise clients.
    • Integrated AWS and GCP cloud solutions for seamless deployment of web applications, enhancing scalability and performance.
    • Automated connections between hospital lab machines and hospital management systems, eliminating manual data entry and reducing errors.
    • Provided comprehensive troubleshooting and networking solutions, ensuring minimal downtime and maximum efficiency across multiple business environments.
    • Set up hosting solutions for web applications, ensuring secure, reliable, and scalable platforms for businesses.